Hepatology International

https://doi.org/10.1007/s12072-025-10894-4

GUIDELINES

Management of acute variceal bleeding: updated APASL guidelines

Cosmas Rinaldi Adithya Lesmana1,2,3 · Akash Shukla4,5 · Ashish Kumar6 · Shalimar7 · Xiaolong Qi8 ·
Rino Alvani Gani1 · Ze‑Hao Zhuang9 · Abdul Kadir Dokmeci10 · Gin Ho Lo11 · Hitoshi Maruyama12 · Ji‑Dong Jia13 ·
Anand V. Kulkarni14 · Jason Chang15 · Necati Ormeci16 · Gamal Shiha17 · Hasnain Ali Shah18 · Jose D. Sollano19 ·
Sahaj Rathi20 · Tan Soek Siam21 · George K. Lau22 · Rungsun Rerknimitr23 · Ming‑Chih Hou24 · Juferdy Kurniawan1 ·
Guohong Han25 · Amar Mukund26 · Sanjay Saran Baijal27 · Shiv. Kumar Sarin26

Received: 27 March 2025 / Accepted: 31 May 2025

© Asian Pacific Association for the Study of the Liver 2025

Abstract
Acute variceal bleeding (AVB) is a common life-threatening complication of portal hypertension (PHT), having a six-week
mortality of 10%-20%. Major advances in the hemodynamic management, risk stratification, pharmacotherapy, endoscopy
techniques, hemostatic devices and radiological interventions have led to improved management and outcome of AVB patients
in the recent past. Therefore, the APASL Portal Hypertension Working Party, chose a panel of experts, primarily from the
Asia–Pacific region, to identify important developments and controversial areas in the field of AVB. They discussed through
a pre-defined and structured process, advances in the field and proposed updates to the previous APASL AVB guidelines.
These included emphasis on safe transportation, defining time frames for AVB episodes and re-bleeding, reporting of clini-
cal outcomes, optimizing early intervention strategies, pharmacotherapy, medical management, endoscopic therapies, and
salvage modalities, including TIPS and self-expanding metal stents. The current updates also cover variceal bleeding in
special populations and situations, the skill sets required for managing AVB patients, and the research priorities in the field.
The updated guidelines are based on the latest evidence and incorporate emerging trends to provide a contemporary template
for management of AVB in both patients with cirrhosis and non-cirrhotic portal hypertension.

Keywords Gastrointestinal hemorrhage · Cirrhosis · Portal hypertension · Vasoactive drugs · Endoscopy · Varices · Non-
cirrhotic portal fibrosis · EHPVO · CTP score · MELD score · Hepatic encephalopathy · TIPS · Ella-Denis Stent · Shunt
surgery

Introduction significant contributions made by the Baveno VII consen-

sus conference on Portal Hypertension [1] and the recent
Acute variceal bleeding (AVB) is a medical emergency and guidelines published by the American Association for the
is associated with a mortality of 10–20% at 6 weeks [1, 2]. Study of the Liver [2]. In the past, the APASL working party
Over the past decade several advances have taken place in on portal hypertension has published guidelines on acute
the diagnosis and management of AVB. Due to the rapid variceal bleeding, extra-hepatic portal vein obstruction [3],
and often episodic nature of AVB, and geographical vari- non-cirrhotic portal fibrosis [4], and primary prophylaxis of
ations in the expertise and the availability of the requisite variceal bleeding [5].
infrastructure, the design and conduct of good clinical trials For the development of the consensus update on AVB,
for the assessment and management for AVB has remained the experts of the APASL working party identified contem-
challenging. To move the field forward keeping in mind all porary and controversial issues and available evidence on
the above issues, the Working party of the APASL on Portal various aspects of AVB. Experts, predominantly from the
Hypertension undertook to update the management of AVB. Asia–Pacific region, were requested by the working party
In developing these updated guidelines, the working to critically analyze the existing literature and develop
party of APASL was fully aware of, and acknowledged the evidence-based consensus statements and recommenda-
tions on each of these issues. Several on-line meetings to
define and focus the issues using the Delphi method were
Extended author information available on the last page of the article

Vol.:(0123456789)
 Hepatology International

undertaken. Each working group conducted its meetings Definition of ‘active’ bleeding
and prepared its consensus statements. These were sub-
sequently presented to the entire group of experts. State- The definition of ‘active’ bleeding was clarified in the
ments that were unanimously approved by all the experts APASL 2011 guidelines. Active bleeding is a state which
were finalized, drafted and then circulated for review, and is defined endoscopically, when spurting or oozing is seen
were subsequently presented at the APASL single theme from the varix. This discrimination between active and
conference on “Portal Hypertension” at Bali, Indonesia, on inactive ‘acute’ bleeding is important because the progno-
October 13, 2024. Level of evidence (LoE) and grade of sis differs between the two. Significance of active bleed-
recommendations were based on the GRADE system [6]. ing at endoscopy has evolved overtime. It predicts early
A summary of the most important consensus statements rebleeding [15].
which form the guidelines along with the background lit-
erature are presented here.

Definition of control of AVB

Definitions of acute variceal bleeding (AVB) Control of AVB is defined as the cessation of bleed-
ing with hemodynamic stability (systolic blood pres-
Variceal bleeding constitutes 70% of all upper gastrointes- sure > 90 mmHg and MAP ≥ 60 mmHg) for at least 48 h
tinal bleeding episodes in patients with portal hyperten- post-therapy​. This definition has remained acceptable over
sion, and it can result from esophageal varices (EVs) and/ the years due to its clarity and utility in clinical settings.
or gastric varices (GVs) [7]. The remaining 30% could be
due to portal hypertensive gastropathy, gastric antral vas-
cular ectasia, peptic ulcer disease (PUD), ectopic varices,
Mallory-Weiss lesions, etc.[8] Definitions of failure to control AVB
AVB needs to be differentiated from recent bleeding and refractory bleeding
and re-bleeding, because the prognosis and management
algorithms differ in these situations. It is important that the In cases of active bleeding confirmed by endoscopy, ces-
defined criteria should be easy, convenient, widely appli- sation of bleeding should be achieved by the end of the
cable for clinical practice and should be able to stratify procedure. Failure to control AVB should be defined as
patients accurately. Currently, there are differences in the persistent bleeding at the end of endoscopy or refractory
definitions relating to AVB. For example, studies for the bleeding despite appropriate therapy, irrespective of the
efficacy of vasoactive agents have used a time frame of 2 time frame, if the clinical situation meets one or more of
to 5 days to define AVB [9]. Those relating to the impact the following criteria:
of infection following AVB have used a time frame of
5–7 days, and those relating to impact of endoscopic thera- 1. Persistence of bleeding at the end of the initial endo-
pies have used a time frame of days to weeks. scopic procedure, as defined by:
We propose that AVB be defined as hematemesis within • Active bleeding from varices despite endoscopic inter-
last 48 h of presentation, and/or ongoing melena, with last vention, or
melanic stool within last 24 h in a known or suspected • Inability to perform endoscopic variceal ligation (EVL)
case of portal hypertension (PHT). The time of the first or other endoscopic interventions due to inadequate
bleed (symptom onset) is considered as ­T0. Recent bleed visualization caused by ongoing bleeding.
refers to a clinically significant bleed which occurs within 2. Refractory bleeding is bleeding which is continuing even
6 weeks of presentation after the episode of AVB. A after appropriate combined pharmacotherapy and endo-
variceal bleed which occurred more than 6 weeks of pres- scopic therapy and is defined as:
entation should be considered as a past bleed. • fresh hematemesis after initiation of combination ther-
Any bout of hematemesis from ­T0 to 48 h of ­T0 should apy (vasoactive drugs and endoscopic treatment), or
be considered as part of the same episode of AVB. The • Drop in hemoglobin of ≥ 2 g/dL or in hematocrit
time frame for AVB as defined in Baveno II[10, 11] and of ≥ 6%, not explained by hemodilution or transfusion
Baveno III[12, 13] was 48 h, but was increased to 120 h adjustments, or
in Baveno IV [14]. However, the expert groups felt that a • development of hemodynamic instability requiring
time frame longer than 48 h will make it difficult to dif- urgent resuscitation (SBP < 90 mmHg, HR > 110 bpm),
ferentiate between early re-bleeding and failure to control or
bleeding.
Hepatology International

• death due to uncontrolled variceal bleeding. Clinically significant re‑bleeding

Clinically significant re-bleeding is defined as a rebleeding

Re‑bleeding and subtypes with a decrease of at least 2 g/dL in hemoglobin with-
out transfusion or an adjusted blood requirement index
Re-bleeding is defined as bleeding occurring after the con- (ABRI) ≥ 0.5 at any time point [16]. This formula aims to
trol of AVB and is classified into three categories: provide a quantifiable measure that reflects the efficiency
of blood transfusion in improving the patient's hematocrit.
• Very early re-bleeding: Rebleeding within 48 to 120 h This definition is retained to guide therapeutic decisions
from endoscopic therapy. based on the clinical impact of re-bleeding.
• Early re-bleeding: Rebleeding between 120 h and 42 days
from endoscopic therapy.
• Late re-bleeding: Rebleeding beyond 42 days from endo-
scopic therapy. Index bleeding
This categorization facilitates better prediction and man- Index bleeding refers to the very first episode of UGI
agement of bleeding recurrence based on the timing of the bleeding.
event (Tables 1, 2 and Fig. 1).

Table 1  Time dependent State Time frame to ­T0 Sub-types Time frame from T
­0
definition of acute variceal
bleeding and re-bleeding Acute variceal bleeding 48 h Active (based on endoscopy) 48 h
Re-bleeding Beyond 48 h Very early rebleed 48–120 h
Early rebleed 120 h—42 days
Late rebleed After 42 days

Table 2  Comparison of time Aspect Baveno VII (2021) APASL 2025

frames: Baveno VII vs. APASL
2025 Duration of an AVB episode 120 h (5 days) 48 h
Definition of re-bleeding Any bleeding after 5 days Any bleeding after 48 h
-Very early re-bleeding Not separately defined 48–120 h (2–5 days) after
AVB control
-Early re-bleeding Beyond 5 days to 6 weeks Beyond 5 days to 6 weeks
-Late re-bleeding Beyond 6 weeks Beyond 6 weeks

Fig. 1  Time frames of acute

variceal bleeding
 Hepatology International

Outcomes in relation to AVB entire clinical course, including pre-hospital care (Fig. 1).
This change aims to improve tracking and evaluation of
To assess the prognosis and clinical outcome of AVB, we define early intervention strategies. Hematemesis was used as
outcome measures in two categories: conventional outcome the reference point for symptom onset. Using the onset
measures (retained from APASL 2011) and newly introduced of melena as time zero in defining acute variceal bleeding
parameters that provide a more comprehensive assessment. is limited by its subjective nature, uncertain timing, lim-
Mortality related to AVB includes all-cause mortality ited patient knowledge and poor specificity. Melena may
within 42 days of the AVB event. appear long after bleeding begins, persist despite hemo-
stasis, and is not exclusive to variceal sources, potentially
New outcome measures introduced to assess delaying diagnosis, and appropriate intervention. Objec-
clinical significance of AVB tive indicators like hematemesis or endoscopic findings
are more reliable for establishing the onset.
The experts deliberated on the emergency management steps
and protocols for AVB. They proposed to include the following
additional outcome measures to be recorded and reported for
standardization of protocols specially in relation to interven- ‘Home‑to‑door’ time
tions. The objectives are to reduce the morbidity, minimize the
complications, and improve survival of patients with AVB: ‘Home-to-door’ time is a novel concept and is defined as
the time elapsed from the onset of bleeding at home to the
• Duration of ICU stay—indirectly captures the severity of time of arrival at the hospital. The ideal time for patients
AVB and likely clinical outcomes post-AVB. to reach to the hospital is proposed to be ideally two hours
• Duration of hospital stay—reflects bleed related events and but may be acceptable for up to 4 h in mild cases. This
healthcare resource utilization. aligns with the “golden hour” principle in emergency care​
• Quality of life—this was evaluated using standardized . Delays beyond 4–6 h are likely to be associated with
assessment tools to gauge patient-reported outcomes post- increased mortality due to uncontrolled bleeding, hepatic
AVB. ischemia, and organ failure.
• New decompensation events—includes new-onset ascites, The updated definitions of AVB and its associated out-
jaundice, hepatorenal syndrome (HRS), or hepatic enceph- comes aim to reflect current knowledge base, clinical prac-
alopathy (HE). tices and incorporate insights from recent studies from dif-
• Delta-MELD Score at 42 Days—change in the MELD ferent emergency scenarios from other branches of medicine.
score from baseline to 42 days post-AVB, indicating wors-
ening of clinical condition or recovery.
• Development of ACLF—identification of acute-on-chronic
liver failure (ACLF) triggered by AVB, as per the APASL Consensus statements
criteria.
• Further decompensation—worsening of pre-existing 1. Definition of acute variceal bleeding and related states
decompensation status of liver disease with progression to 1.1. Acute Variceal Bleeding (AVB) is defined as hemate-
new/additional organ failures beyond the initial bleeding mesis within 24 hours of presentation and/or ongoing
event melena with the last melanic stool within the last 24
hours in a known or suspected case of portal hyperten-
These additional outcome measures were incorporated for sion. (LoE-High; Recommendation-Strong)
reporting following AVB, to enhance the assessment of the 1.2. AVB episode is considered to last for 48 hours from the
long-term prognosis and recovery post-AVB, offering a more first episode of bleeding (­ T0). Any subsequent bleed-
comprehensive patient-centered evaluation beyond conven- ing within 48 hours is part of the same AVB episode.
tional bleeding control and mortality endpoints. (LoE-Moderate; Recommendation-Strong)
1.3. Active bleeding in the context of AVB is defined as:
(LoE-Moderate; Recommendation-Strong)
Definition of time zero ­(T0)

T0 was previously defined as the time of hospital presen- • Endoscopic Findings: Visible spurting or oozing of
tation. However, we propose redefining T ­ 0 as the time of blood from a varix during endoscopy.
first bleed (symptom onset), as this approach captures the • Fresh blood in the nasogastric tube
Hepatology International

• Hematemesis of fresh blood • Control of AVB

• Failure to control AVB
• Re-bleeding events (categorized into very early, early,
1.4 Control of AVB is defined as: (LoE-Moderate; Recom- and late)
mendation-Strong) • Mortality related to AVB (within 42 days)
• Duration of ICU stay.
• Duration of hospital stay.
• Cessation of bleeding with hemodynamic stability for • Quality of life deterioration.
24 h after therapy. • New decompensation, such as ascites, new onset jaun-
• In patients with active bleeding on endoscopy, control dice, acute kidney injury (AKI), HE, sepsis, ACLF, etc.
of AVB indicates cessation of bleeding at the end of • Development of ACLF
the procedure. • Further decompensation
• Development of Ischemic hepatitis
• Delta-MELD score at 42 days post-AVB compared to
1.5. Failure to control AVB is defined by any of the follow- pre-AVB
ing events within 48 hours: (LoE-Moderate; Recom- • Requirement of alternative/rescue therapies; TIPS,
mendation-Strong) PARTO (plug-assisted retrograde transvenous oblitera-
tion), etc.

• Fresh hematemesis despite adequate implementation of

combination therapy (vasoactive drugs and endoscopic 1.10. Time zero ­(T0) is defined as the time of first bleed
therapy). (symptom onset). (LoE-Moderate; Recommendation-
• A drop of ≥ 2 g/dL in hemoglobin or 6% drop in hema- Weak)
tocrit if no transfusion is administered.
• Death
1.11. Home-to-Door Time is defined as the time from symp-
tom onset to hospital arrival. The ideal time is pro-
1.6. Re-bleeding is classified into three categories: (LoE- posed to be 2 hours in most cases, but 4 hours may
Moderate; Recommendation-Strong) be acceptable in less severe cases. (LoE-Low; Recom-
mendation-Weak)

• Very Early Re-bleeding: within 48 to 120 h from ­T0.

• Early Re-bleeding: between 6 and 42 days from ­T0
• Late Re-bleeding: after 42 days Diagnosis, evaluation, and severity
assessment of patients with AVB

1.7. Clinically significant re-bleeding is defined by: (LoE- Patients presenting with UGI bleeding (UGIB) must be
Moderate; Recommendation-Strong) promptly assessed to initiate appropriate management. This
necessitates a more focused evaluation based on key high-
yield indicators in the emergency department. Traditional
• A decrease of 2 g/dL of hemoglobin if no transfusion is detailed clinical examination with features like spider angi-
given, ABRI (adjusted blood requirement index) ≥ 0.5 omas, testicular atrophy etc. may be deferred until urgent
at any time point. hemodynamic measures have been initiated. A quick and
efficient history, physical examination, and initial labora-
tory values are important in assessing resuscitation require-
1.8. Index bleeding is defined as the first episode of UGI ments, triage, endoscopy timing, consultation requirements,
bleeding (LoE-Moderate; Recommendation-Strong) and prognostication [17].
The initial evaluation of a patient with suspected UGIB
begins with a history and physical examination. In par-
1.9. Outcomes of AVB should include: (LoE-Moderate; ticular patients should be asked about previous episodes
Recommendation-Strong) of UGIB, a comprehensive review of recent and current
 Hepatology International

medications, and history of alcohol, tobacco, and sub- Table 3  Imaging signs for cirrhosis or portal hypertension [48, 49]
stance use. All patients should specifically be asked about
Morphologic imaging signs
intake of non-steroidal anti-inflammatory drugs (NSAIDs), Liver parenchyma heterogeneity: liver fibrosis
anticoagulants, anti-platelet agents, and selective serotonin Size of the liver: enlargement and later reduction
reuptake inhibitors because these medications increase the Dysmorphic liver:
risk of bleeding. The goal of the patient history is to iden- Hypertrophy of the left lobe—atrophy of the right lobe
Segment I hypertrophy
tify risk factors that may point to an underlying etiology Segment IV atrophy
of the UGIB. For example, a patient with NSAID use for Enlargement of hilar periportal space
osteoarthritis presenting with UGIB may have peptic ulcer Right posterior hepatic notch sign
disease related bleed, whereas a patient with alcohol abuse Nodularity of liver surface
and cirrhosis may have esophageal varices. Signs of portal hypertension
Portal vein diameter > 12 mm
The history should also include significant co-morbid Spleen length > 11.2 cm
conditions, and whether there is a past history of hepa- Portal velocity max < 18 cm/s, mean velocity < 10 cm/s
titis B or C infection, metabolic dysfunction associated Portosystemic collateral vessels
fatty liver disease (MAFLD) [18], portal hypertension Other hemodynamics changes
or cirrhosis. On examination, important features include Arterial hepatic flow changes
Increase in diameter
jaundice, ascites, hepatic encephalopathy, splenomegaly, Resistive index (RI) > 0.7, Pulsatility index (PI) > 1.2
palpable firm liver, enlarged left hepatic lobe and pres- Demodulation of hepatic veins on Doppler spectrum
ence of dilated abdominal wall veins. The aforementioned Other signs
signs indicate that the bleeding is likely to be associated Ascites
with portal hypertension. In cirrhosis, about 60% of initial Gallbladder wall thickening
Peribiliary cyst
UGIB is from EVs [19].
Laboratory assessments focus on assessing liver function
and coagulation status [20, 21]. The Child-Turcotte-Pugh
(CTP) and Model for End-Stage Liver Disease (MELD) of cohort studies concluded that CECT imaging is superior
scores are utilized to assess the severity of liver dysfunc- to liver stiffness measurement (LSM) and MRI in diagnos-
tion in cirrhosis [20]. Among laboratory markers, throm- ing esophageal varices and predicting cirrhosis patients at
bocytopenia, elevated INR and bilirubin and low albumin high-risk of bleeding [38].
should alert the provider to the possibility of cirrhosis and Non-variceal bleeding in patients with UGIB should be
portal hypertension [21–24]. Ultrasound imaging may reveal suspected if there is abdominal pain or history of NSAID
features of cirrhosis or portal hypertension like an irregu- use, anticoagulants, or antiplatelet agents, past history of
lar liver margin or coarse liver echotexture, ascites, sple- peptic ulcers, Helicobacter pylori infection, as well as the
nomegaly, dilated portal vein, portosystemic collaterals, or history of peptic esophageal lesions, neoplastic lesions,
recanalized paraumbilical vein. Evidence from a retrospec- Dieulafoy’s lesions, angiodysplasia or esophagitis [39–42].
tive study involving 1729 patients with cirrhosis demon- Additionally, symptoms such as retrosternal chest dis-
strated that approximately 60% had spontaneous porto-sys- comfort, epigastric pain, belching, nocturnal pain, nausea,
temic shunts. The prevalence of these shunts increased with and vomiting may further suggest non-variceal bleeding
worsening liver function [25]. Development of spontaneous [43]. Endoscopic findings, including ulcers with spurting or
portal shunts is a marker of the progression of portal hyper- oozing blood (Forrest 1a or 1b) or ulcers with non-bleeding
tension [26–28]. A retrospective study found that cirrhosis visible vessels [44–47], provide further confirmation of non-
patients typically exhibit wider paraumbilical veins and a variceal bleeding as the underlying cause.
greater number of portosystemic collateral channels [26, 27].
Another study identified the diameters of the left gastric vein
and its originating vessels as independent risk factors for the Diagnosis of AVB at endoscopy
presence of varices [29]. Several studies have demonstrated
that splenomegaly is a key indicator of portal hypertension The gold standard test to diagnose AVB is upper gastroin-
[30–36]. Splenorenal shunts, dilated left and short gastric testinal endoscopy and the endoscopic diagnostic criteria of
veins, and umbilical vein recanalization may also be seen in AVB are outlined in the previous APASL Guidelines [15].
computed tomography (CT) and magnetic resonance imag- Variceal hemorrhage is defined as bleeding from an EV and/
ing (MRI) [37]. Imaging signs for hepatic fibrosis and cir- or GV confirmed by endoscopy. AVB may be active or inac-
rhosis have been provided in Table 3. tive at the time of presentation. Active bleeding is a state
CT scan is recommended as a non-invasive diagnostic which is defined endoscopically, when spurting or oozing is
tool for assessing esophageal varices. A systematic review seen from the varix [15].
Hepatology International

The European Society of Gastrointestinal Endoscopy   Gastric varices are universally classified according to
(ESGE) guidelines in 2022 recommended that the diagnosis the Sarin classification, which defines: Type-1 Gastro-
of AVB should be classified into esophageal variceal hem- oesophageal varices (GOV1), which extend below the
orrhage and gastric variceal hemorrhage, with a detailed gastroesophageal junction along the lesser curvature of
description of the endoscopic diagnosis of each [50]. the stomach, Type-2 GOV (GOV2) extend below the gas-
troesophageal junction into the gastric fundus, Isolated
• Esophageal variceal hemorrhage gastric varices—Type I (IGV1), which are located only in
  The endoscopic diagnosis of acute esophageal variceal the fundus and with no or small esophageal varices, and
bleeding is made when there is active hemorrhage from Type-2 IGV (IGV2) are located elsewhere in the body or
a varix, or a sign of recent hemorrhage (nipple sign, antrum of stomach (figures 2, 3).
platelet-fibrin plug). An esophageal variceal source of
upper gastrointestinal hemorrhage can also be inferred Bleeding is considered to have arisen from a gastric varix
when there is blood in the stomach with no other source if (a) active bleeding or oozing of blood is seen from a gas-
of bleeding except for esophageal varices. tric varix, (b) a clot or blackish ulcer is seen over the gastric
• Acute gastric variceal hemorrhage varix, or, (c) in the presence of distinct large gastric varices

Fig. 2  Classification of gastric

varices [51]
 Hepatology International

Fig. 3  Algorithm of manage-

ment in patient with acute
gastric variceal bleeding

and absence of esophageal varices and no other cause of recumbency to standing, suggests a loss of 15% or more of
upper gastrointestinal bleeding is detectable [51]. the blood volume. Hypotension is associated with loss of
These endoscopic criteria play a critical role in the timely at least 40% blood volume [52]. Patients in shock typically
and accurate diagnosis of AVB, ensuring that appropriate have a thready, weak pulse, and cold, clammy extremities.
treatments can be administered quickly. The presence of GV Hyperactive bowel sounds are consistent with an UGIB
with red signs, accompanied by clinical signs of UGI bleed- because blood in the proximal gut is an irritant that stimu-
ing such as melena or hematemesis, is also indicator of acute lates peristalsis, whereas normal bowel sounds are more
GV bleeding. Classification based on vascular anatomy is consistent with lower gastrointestinal bleeding.
important to decide the type of radiological interventions in The expert panel revisited the predictors of severity of
patients with GVs [51]. When both gastric and esophageal AVB, treatment failure and early rebleeding including the
varices are present and the source of bleeding cannot be APASL Severity Score for AVB assessment (Tables 4,5).
clearly identified; such cases may be classified under the Most early studies showed that alcoholic liver disease
category of ‘variceal bleeding from uncertain source’. This patients have more severe AVB, but recent studies have
terminology, however, needs to be assessed further. suggested better outcomes in alcohol-associated cirrhosis
[53, 54]. Several factors, such as INR, encephalopathy, and
Child–Pugh score overlap significantly. While these vari-
Severity assessment and evaluation ables may not serve as independent predictors, they contrib-
of patients of AVB ute to scoring systems (like, MELD, AIMS65, Neutrophil-
to-Platelet Ratio, Platelet-Albumin-Bilirubin Index) that are
The severity of blood loss is roughly estimated by the hemo- instrumental in determining outcomes in AVB [55].
dynamic status and other key signs. Visible blood loss can be
estimated based on the patient's history and clinical observa-
tions. Resting tachycardia, in the absence of another cause
and maintained blood pressure, suggests mild to moderate Consensus statements
hypovolemia. Orthostatic hypotension, defined as a decrease
in the systolic blood pressure of more than 20 mmHg or 2. Diagnosis, evaluation, and severity assessment of
an increase in the pulse of more than 20 beats/min from patients with acute variceal bleed
Hepatology International

Table 4  Predictors of severity Predictor Severity of Treatment Early re- Mortality

of acute variceal bleeding, variceal bleed failure bleeding
treatment failure, early
re-bleeding, and mortality HVPG Yes Yes Yes Yes
Alcoholic liver disease Yes Yes Yes Yes
Infection Yes Yes Yes Yes
CTP class/score Yes Yes – Yes
PRBC transfusion Yes Yes – Yes
Size and morphology of varices Yes Yes – –
Ascites Yes Yes – –
Portal vein thrombosis Yes Yes – –
Hematocrit/Hemoglobin at presentation Yes Yes – –
Platelet count Yes – Yes –
Degree of liver failure Yes – –
Active bleeding at endoscopy – Yes Yes Yes
Shock – Yes – Yes
AST – Yes – –
First bleed – Yes – –
MELD > 18 – – Yes Yes
Encephalopathy – – Yes Yes
Hepatocellular carcinoma – – Yes Yes
Short interval to admission – – Yes Yes
Blood urea – – Yes Yes
Hematemesis – – Yes
S. creatinine – – – Yes
S. albumin – – – Yes
Age – – – Yes
Early re-bleeding – – – Yes
Prothrombin time – – – Yes
Treatment failure – – – Yes
S. bilirubin – – – Yes

HVPG: hepatic vein pressure gradient; CTP: Child-Turcotte-Pugh; PRBC: packed red blood cell; AST:
aspartate aminotransferase; MELD: Model for End Stage Liver Disease

Table 5  APASL bleed severity Parameter Value Point

score for acute variceal
bleeding* Systolic blood pressure > 90 mmHg and no postural drop 0
> 90 mmHg with postural drop 1
< 90 mmHg 2
Child-Turcotte-Pugh class A 0
B 1
C 2
Platelet count ≥ 100.000 mm-3 0
< 100.000 mm-3 1
Infection Absent 0
Present 1
Active bleeding at endoscopy Absent 0
Present 1
Total Minimum 0,
maximum
7
*
Reproduced from 2011 APASL guidelines[15]
 Hepatology International

2.1 In a patient with UGI bleeding, if the following are pre- • Presence of a sign of recent bleed over a GV (overlying
sent, suspect the cause to be variceal bleeding: clot or white nipple sign)
• Presence of GV with red signs (risk factor for bleeding)
and presence of blood in the stomach in the absence of
• Previous history of hepatitis B and C, or alcohol abuse, another source of bleed/or stigmata of recent bleed on
or metabolic-associated fatty liver disease [18], or esophageal varices
exposure to hepatotoxic substances [56, 57], or long-
term use of hepatotoxic medications [48, 58] (LoE-
Moderate; Recommendation-Strong) 2.4. For describing site of acute GV bleeding, Sarin’s clas-
• Physical findings: jaundice, ascites, signs of hepatic sification of gastric varices should be used. (LoE-High;
encephalopathy, splenomegaly, firm hepatomegaly, Recommendation-Strong)
abdominal wall collaterals or signs of hepatic failure.
(LoE-Moderate; Recommendation-Strong)
• Imaging signs: spontaneous portosystemic shunts, 2.5. For describing site of acute GV bleeding, Sarin’s clas-
dilated left and short gastric veins, umbilical vein reca- sification of gastric varices should be used. (LoE-High;
nalization, splenomegaly, nodular aspect of the liver, Recommendation-Strong)
segmental dysmorphia, hypertrophy and hypotrophy
of the liver, and enlargement of hilar periportal space.
(LoE-Moderate; Recommendation-Strong) • Change of vital signs (heart rate, blood pressure)
• Laboratory data: thrombocytopenia, elevated INR, hepatic • Hematocrit/hemoglobin
synthetic dysfunction, albumin and bilirubin abnormalities. • Transfusion requirement
(LoE-Moderate; Recommendation-Strong) • Adjusted Blood Requirement Index
• Endoscopy: signs of variceal bleeding. (LoE-High;
Recommendation-Strong)
2.6. Predictors of severe acute variceal bleeding and
rebleeding should be carefully assessed at presenta-
2.2. The gold standard for diagnosis of acute variceal bleed- tion (Table 4). (LoE-High; Recommendation-Strong)
ing is UGI endoscopy. (LoE-Moderate; Recommenda-
tion-Strong)
2.3. Endoscopic diagnosis for variceal bleeding 2.7. The APASL bleed severity score is a simple and reli-
able tool for assessing the severity of acute variceal
bleeding and should be used to assess severity of
2.3.1. Acute esophageal variceal bleeding (LoE-High; AVB. (Table 5) (LoE-Moderate; Recommendation-
Recommendation-Strong) Strong)

• Direct visualization of blood arising from an esopha-

geal varix—usually spurting or oozing. Resuscitation, initial management,
• Presence of a sign of recent bleed on a varix (white and monitoring of patients with acute
nipple sign or overlying clot. variceal bleeding
• Presence of esophageal varices with red signs (risk fac-
tor for bleeding) and presence of blood in the stomach Management in pre‑emergency period
in the absence of another source of bleeding.
• Presence of esophageal varices with red signs and clini- The management of AVB includes protection and main-
cal signs of UGIB without blood in the stomach. tenance of airway, breathing, and circulation. Patients
with AVB need urgent transport using specially equipped
advanced life-care ambulance. It is very important that
2.3.2. Acute gastric variceal bleeding (LoE-High; Rec- transport crew understands the underlying pathophysiology
ommendation-Strong) of variceal bleeding. A quick history and examination of
patient by the transport team is useful to suspect PHT. A
history of using direct oral anticoagulants or anti-thrombotic
• Direct visualization of blood arising from a GV, spurting or anti-platelet medication need prioritization for transport
or oozing because the high risk of persistent and severe bleeding.
Hepatology International

Ultrasound by the transport crew can help discriminate pos- placed in the emergency room, if there is a strong suspicion
sible variceal and non-variceal UGIB. of bleeding or associated hepatic encephalopathy [67].
Airway intubation is indicated in patients who are bleed-
ing severely, who have mental status changes, and difficult Pharmacotherapy in AVB
to maintain oxygen saturation above 90%. Hemodynamic
condition should be monitored by using pulse oximeter and Antibiotics must be promptly initiated in patients with AVB.
continuous cardiac monitoring during transportation. Vas- Ceftriaxone is the preferred antibiotic due to its coverage of
cular access needs to be placed for volume resuscitation and gram-negative bacteria, the commonest infection causing
hemodynamic support. Two large bore (≥ 18 G) peripheral bacteria in this setting. Ceftriaxone was found to be superior
intravenous lines should preferably be placed. Ultrasound to norfloxacin in preventing infections in patients with cir-
guided IV placement, or central venous access may be war- rhosis in a RCT [68]. In this trial of 111 patients with cir-
ranted in patients with difficult peripheral access. Transport rhosis and GI bleeding, those receiving oral norfloxacin had
crew, should preferably be trained for emergency balloon higher infection rates (26% vs. 11%, p = 0.003) and sponta-
tamponade for hemodynamically unstable patients or those neous bacterial peritonitis (12% vs. 2%, p = 0.003) compared
with clinical features of ongoing bleed. The systolic blood to those given intravenous ceftriaxone [69]. Key factors in
pressure should be maintained at least at 90–100 mmHg, and selecting antibiotics include individual patient characteris-
the heart rate should be maintained below 100 beats/ min. tics (e.g., prior antibiotic exposure or infection) and local
Terlipressin bolus of 2 mg, can be given in the ambulance antibiotic resistance pattern [69, 70].
or during transport of the patient (pre-ER). Upper GI bleeding in a patient with chronic liver dis-
Artificial intelligence (AI) can be used as a tool for patient ease or cirrhosis should be deemed from GEVs and thus
risk assessment pre-endoscopy period. This can increase vasoactive treatment (terlipressin or somatostatin/octreotide)
segregation of low-risk patients who can be safely dis- should be started at the earliest, preferably within 30 min
charged for after initial management at the hospital [59]. The of index bleed [71]. This should be followed by endoscopy
AI can also be used for predicting the risk for re-bleeding. and the appropriate treatment [72]. Terlipressin, somatosta-
Study from Levi et al. showed that the machine learning tin or octreotide are all vasoconstrictors which can be used
algorithm can offer reliable guidance (AUC > 0.80) for pre- in AVB. The choice among the vasoconstrictors may be dif-
dicting the need for transfusion in the next 24 h.[60] ficult and often depends upon the availability, affordability
and patient co-morbidities [73]. There are greater risks of
Management in the emergency room arrhythmias with the use of terlipressin and therefore must
be used with caution or avoided in patients who are at risk
The principles to protect airway and breathing are the like IHD or atrial fibrillation. On the other hand, terlipressin
same between ER and pre-ER management. It is important should be preferred in the presence of AKI [74, 75]. The effi-
to identify AVB patients at high risk for failure to control cacy in the initial control of variceal bleeding is comparable
bleeding or requiring emergency intubation. Prophylactic amongst the three vasoactive agents [76]. Duration of phar-
endotracheal intubation in the setting of UGIB can be asso- macological treatment should be kept for the optimal time
ciated with higher rates of respiratory complications [61]. with a minimum of 2 days, the duration for AVB [77–79].
Circulation needs to be monitored, with a target hemoglobin Terlipressin infusion (4 mg/24 h) is preferred over the IV
level around 7–8 g/dL (hematocrit of 21–24), because equal bolus injections, as the former modality achieves higher
or over transfusion can cause a rebound increase in portal success rates in the control of AVB with lower dosages and
pressure and precipitate early re-bleeding [62, 63]. Patients fewer complications [80].
with co-morbidities, such as coronary artery disease need to The use of PPIs in acute UGIB due to non-variceal etiol-
maintain a higher hemoglobin to maintain cardiac perfusion. ogy is a bit controversial. While, PPIs can reduce the inci-
Empiric correction of coagulation parameters with fresh fro- dence of post-EVL ulcers, they do not reduce the number
zen plasma (FFP) and/or platelet transfusion is associated of ulcers, severity of symptoms, and duration of hospital
with worse outcomes in AVB due to the increased rebleed- stay [81–84]. Some studies suggest that the duration of PPI
ing [64, 65]. Thromboelastography (TEG) or Rotational use should be at least 10 days or one month after stopping
Thromboelastometry (ROTEM) can be used for guiding the bleeding [80, 85]. A single study showed that a short
correction of coagulation in patients with difficult to con- course of vonoprazan 20 mg/day is safer and superior to
trol bleed [66]. Echocardiography should be used to assess pantoprazole 40 mg/day in the reduction of post-EVL ulcers
fluid status and cardiac function. Fluid replacement should and prevention of ulcer-related bleeding [86]. Acid suppres-
be used very conservatively and cautiously. Crystalloids sion is superior to no acid suppression to prevent post-EVL
are preferred and colloids should be avoided, particularly complications [87]. However, the benefits of PPIs in reduc-
dextran solution or albumin. The naso-gastric tube can be ing the variceal rebleeding, specially from the EVL ulcers,
 Hepatology International

is not established. Routine use of sucralfate in AVB should • The volume of fluids should be aimed to maintain: (LoE-
also not be recommended due to lack of data. Moderate; Recommendation-Strong)
Tranexamic acid significantly reduces the failure to con- • Systolic blood pressure of 90–100 mmHg
trol bleeding post-EVL by day 5 and failure to prevent re- • Heart rate below 100 beats per minute
bleeding after day 5 to week 6 in patients with advanced • CVP 1–5 mmHg
liver cirrhosis presenting with UGIB by reducing post-EVL • Diuresis of 40 mL/h
ulcer bleeding. However, the use of tranexamic acid does not
reduce mortality [88]. The use of tranexamic acid should be
carefully considered due to the potential risk of portal vein 3.1.4. Blood volume restitution
thrombosis (PVT). 3.1.4.1. Blood transfusion requirement is determined by
As MAFLD/MASLD and related cirrhosis become more estimating blood loss. (LoE-Low; Recommendation-
prevalent, a growing number of patients on anticoagulation Weak)
therapy are undergoing EVL for acute variceal bleeding. 3.1.4.2. Blood volume replacement should be done cau-
While coagulopathy may warrant correction, endoscopy tiously and conservatively to maintain:
can be safely performed at therapeutic levels of anticoagu-
lation in selected cases, particularly for urgent/emergency
procedures. Clinicians should emphasize individualized risk • A hemoglobin level of approximately 7–8 g/dL, depend-
assessment for thromboembolism and bleeding when plan- ing on other factors, such as patient's co- morbidities,
ning endoscopy [89]. age, hemodynamic status, and presence of ongoing
bleeding.
• Packed red blood cells (PRBC) is the preferred blood
component. (LoE-Low; Recommendation-Weak)
Consensus statements • Fresh frozen plasma (FFP) and platelet transfusion
should be given very cautiously due to its potential risk
3. Resuscitation, initial management, and monitoring of for higher rebleeding rate. (LoE-Moderate; Recommen-
patients with acute variceal bleed dation-Weak)
3.1. Initial resuscitative measures include protection of • Specific management of coagulopathy or thrombocytope-
airway, breathing, and circulation (ABC). (LoE-High; nia needs to be studied further for its relevance in acute
Recommendation-Strong): variceal bleeding management. (LoE-Low; Recommen-
3.1.1. For protection of the airway elective intubation is dation-Weak)
recommended in patients with: (LoE-Moderate; Rec-
ommendation-Strong)
3.2. Pharmacological Therapy
3.2.1. Vasoactive treatment should be started at the earli-
• Massive hematemesis est, preferably within 30 minutes of index bleed. (LoE-
• Altered mental status High; Recommendation-Strong)
• Known gastric varices 3.2.2. Terlipressin, somatostatin, or octreotide can be used
• Difficulty in maintaining oxygen saturation above 90% in the prescribed doses depending upon availability
• Massive ascites and patient comorbidities. (LoE-High; Recommenda-
tion-Strong)
3.2.3. Use of terlipressin requires baseline ECG. (LoE-
3.1.2. Monitor hemodynamic condition preferably with Low; Recommendation-Weak)
cardiac monitor 3.2.4. Terlipressin infusion (4-6 mg/24 hours) is preferred
3.1.3. Fluid volume replacement over intravenous boluses (2 mg/4 hours). (LoE-High;
Recommendation-Strong)
3.2.5.In patients where endoscopic therapy has successfully
• Fluid replacement should be used conservatively and controlled AVB, the vasoactive drugs can be discontinued
cautiously. (LoE-Moderate; Recommendation-Weak) after 24 48 hours, and non-specific beta-blocker (NSBB)
• Crystalloids particularly saline is preferred and colloids may be initiated, if there is no contraindication. (LoE-
should be avoided; maintenance fluids should be plasma- Moderate; Recommendation-Strong)
lyte/ dextrose infusion. (LoE-Moderate; Recommenda- 3.2.6. Tranexamic acid may reduce the risk of post-EVL
tion-Weak) rebleeding. (LoE-High; Recommendation-Strong)
Hepatology International

3.2.7. For patients with cirrhosis with AVB, a broad- Table 6  Checklist to be Maintained in the Endoscopy Room
spectrum antibiotic like ceftriaxone (1-2 gram daily)
Patient
is recommended for 5 days before endoscopic therapy. Vital signs
However, the duration could possibly be shortened to Two intravenous lines
2 days in patients with successful endoscopic therapy. Judicious fluid/blood resuscitation
(LoE- High ; Recommendation-Strong ) Supplemental oxygen
Child-Pugh status
3.2.8. In patients with Child-Pugh A cirrhosis, benefit of Informed consent
antibiotics is not evident.[67, 90] (LoE-High; Recom- Endoscopy room
mendation-Weak) Check endoscope (air–water channel, suction, knobs, etc.)
3.2.9. The use of PPI in patients with acute UGIB due to Suction device
non-variceal etiology is helpful. (LoE-High; Recom- Patient resuscitation chart
Patient monitor
mendation-Strong) Accessories
3.2.10. The use of factor VIIa is not routinely recom- Intubation facility with anesthetist support and availability of alter-
mended. (LoE-High; Recommendation-Weak) nate therapy
Sengstaken-Blakemore tube
Esophageal variceal self-expanding metal stent (SEMS)
Interventional radiology services
3.3. Patients with active variceal bleeding should be man- Gastrointestinal surgical team
aged in ICU. (LoE-Low; Recommendation-Weak)
3.4. Monitoring
3.4.1. Naso-enteric tube can be used if there is a strong numbers of patients needed to treat (NNT) 8 and 5, respec-
indication, such as hepatic encephalopathy. (LoE-Low; tively. The difference in favor of combined treatment
Recommendation-Weak) remained significant when trials that used drugs other than
3.4.2. Cardiac monitoring and IVC monitoring are help- octreotide or that included a low proportion of alcoholic
ful to optimize the decisions concerning fluid replace- patients (< 40%) or high-risk cirrhotic patients (< 35%) were
ment, specially in: (LoE-Moderate; Recommendation- excluded. Mortality was not significantly decreased by com-
Strong) bined therapy (RR 0.73, 95% CI 0.45–1.18) [93]. To ensure
smooth conduct of endoscopy in these patients a checklist
should be maintained in the endoscopy room (Table 6).
• Elderly
• Patients with cardiovascular co- morbidity
• Active bleeding at endoscopy Timing of endoscopy
• Patients with severe bleeding
• Presence of shock Variceal bleed can precipitate decompensation and infec-
• Renal failure (impending or present) tions or even exsanguination or hemorrhagic shock. How-
ever, very early endoscopy does not appear to improve out-
comes; However, the data is heterogenous and has provided
Role of endoscopy in AVB conflicting results [89–93]. Endoscopy may be done as soon
as the patient is stable enough for the procedure and logistics
Endoscopic variceal ligation (EVL) is more effective than allow, preferably within 12 h and, at the latest, within 24 h. If
endoscopic variceal sclerotherapy (EST) with greater control the patient is hemodynamically unstable after resuscitation
of hemorrhage, lower re-bleeding rate, and lower adverse or is actively bleeding, endoscopy should be performed as
events but without differences in mortality [91, 92]. Endo- soon as possible [94–97]. There are three proposed timelines
scopic therapy and medical therapy with vasoactive drugs that require further research: < 6 h as urgent, 6–12 h as early,
have a synergistic effect as their modes of action are com- and > 12 h as delayed.
pletely different. A meta-analysis by Banares and co-work-
ers, who compared endoscopic therapy with combined endo-
scopic and pharmacologic treatment, showed that the control Bedside vs endoscopic room
of AVB was more often achieved with combined treatment
than after endoscopic treatment alone. Eight trials involv- Endoscopy should preferably be performed in the well-
ing 939 patients were included in the meta-analysis. Com- equipped endoscopy room whenever possible, ensur-
bined treatment improved initial control of bleeding [relative ing that all hemostatic accessories are available and the
risk (RR) 1.12, 95% confidence interval (CI) 1.02–1.23], option to change scopes in case of channel blockage and
and 5-day hemostasis (RR 1.28, 95% CI 1.18–1.39), with the need of EUS is available. However, patients who are
 Hepatology International

hemodynamically unstable and/or require airway protection, Sedation principles in AVB

should be scoped at the bedside in the ICU.
Sedation with propofol and midazolam is generally safe
and is associated with relatively lower procedure time.
However, no impact on outcomes has been observed [103].
Expertise of endoscopists (attending In patients requiring longer endoscopy procedures, (inject-
or fellow) ing glue or EUS), propofol should be preferred over mida-
zolam as it is safer with low procedure and recovery time.
There is no systemic data explaining who should perform the
therapeutic endoscopic procedures in a patient with AVB.
In most cases, band ligation of varices is a relatively low
complexity procedure and can be performed by a fellow in
Prokinetics
a supervised setting. Glue injection for bleeding GV can be
The use of prokinetics prior to endoscopy improves
a challenging procedure with a steeper learning curve. It
mucosal visualization and reduces the need for second-
should preferably be done by an attending/consultant/faculty
look endoscopy. Erythromycin 250 mg or metoclopramide
or a well-trained fellow under supervision. Endoscopy for
10 mg can be administered 30–120 min prior to endoscopy
patients who are hemodynamically unstable or those with
[104, 105]. There is no data available regarding the use of
major co-morbidities should be done by an attending/con-
simethicone prior to endoscopy.
sultant/faculty to reduce the scope time. Longer scope time
If no bleeding source is visible and fundus is obscured
may increase pulmonary complications [98].
by blood clots, the patient can be moved to a semi-recum-
bent position for better visualization of the fundus [106].
If stomach is clean and no varices are visible, then patient
is managed as obscure overt GI bleeding. Colonoscopy
Type of scope and equipment needed
and/or CT angiogram should be done to rule out other
causes such as hemosuccus pancreaticus or telangiectasias.
Normal UGI scope is generally adequate to manage most
Varices may be decompressed after a bleed and hence a
AVB patients. Rarely, if active bleeding continues, a thera-
relook endoscopy could be done after 24–48 h in patients
peutic gastroscope is preferred due to a larger suction chan-
with index of suspicion.
nel, and facility to pass desired accessories. One should have
other accessories readily available, such as ligators, 21 and
23 G sclerotherapy needles, preferred type of tissue adhe-
sive glue (N-butyl-cyanoacrylate or N-octyl-cyanoacrylate), Recommended endoscopic therapy
hemostatic clips including over the scope clips, hemostatic for esophageal variceal bleeding
powder, sclerosant and electrocautery setup with accesso-
ries—hemostatic cautery-enabled forceps, argon plasma Endoscopic variceal band ligation (EVL) is the preferred modal-
coagulation catheters, snare for clot removal or spray coagu- ity of therapy for esophageal variceal bleed. Endoscopic sclero-
lation should also be available. therapy is discouraged due to higher rates of rebleed and adverse
events. It may be considered in patients with variceal bleed in the
setting of post-banding ulcer/fibrosis [107–109], or in children
where passage of band ligator devise may not be possible. If
Intubation / extubation both gastric and esophageal varices are present in a patient with
variceal bleed, and esophageal banding is planned, it is advisable
Patients with variceal bleed are at a high risk of pulmonary to obliterate GV first, even in the absence of stigmata of recent
infections, possibly due to aspiration [99]. Routine intuba- hemorrhage on GV.
tion and nasogastric tube placement should be avoided due
to the increased risk of pulmonary infections [100–102].
Intubation should be restricted to patients who are actively
Role of EUS
bleeding, and in patients with encephalopathy who are
unlikely to tolerate endoscopy without sedation. If intuba-
EUS can be used to assess esophageal variceal anatomy
tion is performed, extubation should be carried out as early
and predictors for variceal recurrence and is often used
as possible after successful completion of the procedure.
to directly inject glue into esophageal or gastric varices
[110], and ectopic varices [111].
Hepatology International

GAVE‑related bleeding EVL-induced ulcer bleeding can be done using Jamwal &
Sarin classification. The post-EVL ulcer is classified based
Gastric antral vascular ectasia (GAVE) is an uncommon but on their endoscopic appearance. In type A, there is active
important cause of upper gastrointestinal bleeding that is dis- spurting from the ulcer. Type B is characterized as active
tinct from variceal bleeding. It is typically seen in patients with oozing from the ulcer. In Type C, the ulcer appears with
chronic liver disease, systemic sclerosis, diabetes, autoimmune pigmented base, or a visible clot and the Type D ulcer has
disorders, or chronic renal failure. Unlike portal hypertensive a clean base (white or yellow) [114].
gastropathy (PHG), GAVE-related bleeding occurs due to There is no evidence demonstrating the superiority of
abnormal dilated capillaries in gastric mucosa and lamina one treatment technique over another. Treatment options
propria in the gastric antrum, leading to persistent occult or include hemostatic agents, injection of glue or sclerosant,
overt bleeding. Management is primarily endoscopic, with fully covered esophageal metallic stent, and EUS-guided
APC being the first-line therapy, as it effectively coagulates vascular therapy. Combined therapy using one or more
superficial vascular ectasias. Alternative modalities include endoscopic techniques can be considered in difficult to
spray coagulation (hemostatic powders) for diffuse GAVE control bleeding. TIPS is an effective rescue therapy [115].
and endoscopic band ligation for focal nodular lesions, which
may be preferable in cases of chronic or refractory bleeding.
In severe or recurrent cases, radiofrequency ablation (RFA)
and surgical antrectomy may be considered, though these are Endoscopic therapy, recommendation
rarely required. Since GAVE-related bleeding is not related for gastric variceal bleeding
to portal hypertension, its management is distinct from acute
variceal bleeding and should be approached separately in clini- Bleeding from GVs is generally more severe than bleeding
cal decision-making. from EVs but is thought to occur less frequently. Gastric
varices in the fundus are GOV2 and IGV1, with the high-
Role of relook‑endoscopy once successful est bleeding and re-bleeding rates [51]. Large IGV1 may
hemostasis is achieved bleed despite HVPG values less than 12 mmHg [116, 117].
Because the blood flow in the GVs is relatively large and
Limited data suggests no benefit of relook-endoscopy once the bleeding is rapid and often profuse, endoscopic means
successful hemostasis is achieved, and therefore, it is not rou- of treating bleeding GVs are the treatments of choice. The
tinely recommended. Relook-endoscopy may be considered choice of endoscopic therapy used often depends on local
if endoscopist is not certain of complete hemostasis, clini- availability and expertise. Endoscopic cyanoacrylate glue
cal signs of rebleed (new onset of hematemesis, new drop injection is the most preferred procedure for bleeding gas-
in hemoglobin, and new onset hemodynamic instability) tric varices [118]. Other therapies to control hemorrhage
(Table 8) [112, 113]. Melena may persist for 3–5 days after include radiological options (TIPS, balloon-occluded ret-
hematemesis, and in isolation, should not be taken as indica- rograde transvenous obliteration (BRTO). Uncontrolled
tion for check endoscopy. Drop in hemoglobin is defined as data comparing these therapies in bleeding gastric varices
progressive decrease in Hb for more than 1 g/dL or more. show that the best control of initial hemorrhage (90–100%)
is achieved with glue, TIPS, or balloon-occluded retro-
grade transvenous obliteration [119]. In three small single-
Post‑banding drugs center RCTs, endoscopic variceal obturation (EVO) with
glue versus EVS[120] or EVL in bleeding gastric varices
Vasoactive drugs may be continued for 48 h after success- were compared [121, 122]. All three RCTs reported higher
ful endoscopic hemostasis for varices. Non-specific beta- efficacy of EVO in the control of acute hemorrhage [120,
blocker (NSBB) therapy may be introduced once vasoactive 121], re-bleeding [122], or complication rate [121]. Unfor-
drugs are stopped or at day 6 after the control of acute bleed. tunately, less than 50% of the patients included in these
studies had GOV2/IGV1 varices, and a separate analysis
was not performed.
Endoscopic management of EVL‑induced It is recommended that TIPS be used in acute bleed-
ulcer bleeding ing from gastric varices when EVO is unavailable or if
re-bleeding occurs after EVO; however, this has not been
Diagnosis of EVL-induced ulcer bleeding should be made evaluated prospectively. A small single-center study com-
based on endoscopy with ooze or spurt or clot and no evi- paring EVO versus TIPS in the prevention of recurrent
dence of other source of bleed. Severity assessment of hemorrhage in patients in whom acute gastric variceal
 Hepatology International

hemorrhage was controlled with EVO showed similar re- • Extubation needs to be performed as early as possible
bleeding rates, but again fewer than 50% of the patients after control of bleeding. (LoE-Moderate; Recommen-
were bleeding from GOV2/IGV1 varices [123]. dation-Weak)

4.6.4. Drugs (LoE-High; Recommendation-Strong)

Consensus statement

4. Role of endoscopy in acute variceal bleeding • A vasoactive drug should be initiated prior to endos-
4.1. All patients with acute UGI bleed should undergo copy.
endoscopy with the intent to provide endo therapy. • The use of prokinetics prior to endoscopy improves
(LoE-Low; Recommendation-Weak) mucosal visualization and reduces the need for second-
4.2 Combination of a vasoactive drug and endoscopic ther- look endoscopy.
apy is the first-line therapy for acute variceal bleed. • Erythromycin 250 mg or metoclopramide 10 mg can be
(LoE-High; Recommendation-Strong) administered 30–120 min prior to endoscopy.
4.3 Timing of endoscopy (the door -to-scope time): • There is no data available regarding the use of simethi-
4.3.1. In patients with AVB, endo therapy should be done as cone prior to endoscopy.
soon as possible under resuscitation, preferably within 6
hours of admission. (LoE-Low; Recommendation-Weak)
4.3.2. Endoscopy may be undertaken as soon as patient 4.6.5. Sedation (LoE-Low; Recommendation-Weak)
is stable (preferably within 12 hours, latest up to 24
hours). (LoE-Moderate; Recommendation-Weak)
4.4. Place to perform endoscopy • For patient comfort and better visualization, use of seda-
4.4.1. Endoscopy is preferably done in endoscopy room. tion is recommended
Patients who are hemodynamically unstable and/or • Both propofol and midazolam are safe, though the former
need of airway protection, should be scoped at bedside is preferred.
in the ICU. (LoE-Moderate; Recommendation-Weak) • A checklist of activities and required equipment and
4.5. Band ligation of varices can be done by a fellow in a accessories should be maintained in the endoscopy
supervised setting. (LoE-Low; Recommendation-Weak) theater (Table 6).
4.5.1. Skill level of endoscopist
4.5.2. Glue injection for gastric varices is preferably done
by an attending or a well-trained fellow under supervi- 4.6.6. Endoscopic treatment (LoE-High; Recommenda-
sion. (LoE-Low; Recommendation-Weak) tion-Strong):
4.5.3. Endoscopy for patients who are in shock, intubated,   Band ligation with multi-band ligator is the preferred
or bedside in the ICU should be done by a consultant. modality of therapy for esophageal variceal bleed.
(LoE-Low; Recommendation-Weak) 4.6.7. Role of relook- endoscopy is presented in Tables 7,
4.6. Following preparations should be done prior to endos- 8. (LoE-Low; Recommendation-Weak)
copy. (LoE-Moderate; Recommendation-Strong)
4.6.1. Blood pressure: systolic BP > 70 mmHg
4.6.2. Unconscious patients should be intubated prior to
endoscopy. (LoE-Low; Recommendation-Weak) 4.7. Vasoactive drugs
4.6.3. Intubation / extubating (LoE-Moderate; Recom- 4.7.1. Vasoactive drugs may be continued for 48 hours
mendation-Weak) after successful endoscopic hemostasis for varices.
4.7.2. Non-selective beta-blocker (NSBB), like carvedilol
may be initiated once vasoactive drugs are stopped or
• Routine intubation and NGT placement should be after 5 days of acutevariceal bleed.
avoided. (LoE-Moderate; Recommendation-Weak) 4.8. Management of acute GV bleed
• Intubation should be restricted to patients who are 4.8.1. Following on the success in EV bleed, combination
actively bleeding, have encephalopathy, and are of vasoactive drugs and endoscopic therapy should be
unlikely to tolerate endoscopy without sedation. (LoE- used for acute GV bleed, though specific data is lack-
Moderate; Recommendation-Weak) ing. (LoE-Low; Recommendation-Strong)
Hepatology International

Table 7  Indications for

Endotracheal Intubation in AVB Massive hematemesis and active bleeding
Persistent hematemesis with hemodynamic instability (SBP < 90 mmHg, HR > 110 bpm, altered menta-
tion)
Severe hypovolemic shock requiring aggressive resuscitation
High risk of aspiration
Severe hepatic encephalopathy (Grade III-IV) with an inability to protect the airway
Unresponsive or semi-conscious patients with AVB
Gastric retention (gastroparesis, gastric outlet obstruction, massive blood in the stomach) increases the
aspiration risk
Planned therapeutic endoscopy procedures with high aspiration risk
EUS-guided gastric variceal therapy or procedures requiring prolonged sedation
Presence of significant blood in the stomach impairing visualization
Need for prolonged endoscopic intervention where airway protection is necessary
Severe respiratory compromise
Acute respiratory failure, ARDS, or worsening hypoxia
Severe metabolic acidosis (pH < 7.2) or hypercapnia requiring ventilatory support

Role of rescue therapies in AVB

Table 8  Indications of relook-endoscopy
Rescue therapy is indicated when endoscopic treatment or
1 Endoscopist not certain of complete hemostasis combination treatment has failed to control bleeding. Bal-
2 Clinical sign of rebleed loon tamponade using Sengstaken Blakemore tube enables
New onset of hematemesis temporary control of bleeding, by direct compression of
New onset or continued Hb drop varices at the esophagogastric junction, in 40–90% of cases.
New onset of hemodynamic instability Owing to high rates of complications and re-bleeding [37,
124], balloon tamponade is not used routinely as the first-
line treatment for control of AVB.
TIPS is a better alternative when available, in the pres-
ence of failure of combined pharmacologic plus endoscopic
4.8.2. Endoscopic cyanoacrylate injection (ECI) is the therapy. In the Baveno recommendations, it was considered
preferred procedure for bleeding GV. (LoE-Moderate; that a second attempt at endoscopic therapy was also an
Recommendation-Strong) option and one could perform TIPS after failure of the sec-
4.8.3. In patients with acute bleeding from GOV1 type of ond endoscopic therapy [1]. There is data to suggest that
varices, treatment should be similar to that of esopha- TIPS placement is associated with a significant improvement
geal varices or glue injection. (LoE-Moderate; Recom- in survival in patients with HVPG greater than 20 mmHg
mendation-Strong) [125]. Therefore, HVPG can provide useful information that
4.8.4. Endoscopic ultrasound (EUS) based glue injection, allows for risk stratification and more aggressive treatment
with or without coil. in high-risk patients.
4.9. Management of portal hypertensive gastropathy related
bleeding
4.9.1. Vasoactive agents are the first line of therapy for The role of interventional radiologist (IR)
portal hypertensive gastropathy (PHG) related bleed- in AVB
ing. (LoE-Moderate; Recommendation-Strong).
4.9.2. TIPS should be considered as a rescue therapy Esophageal varices
for refractory PHG related bleeding. (LoE-Moderate;
Recommendation-Strong). Pre-emptive TIPS with PTFE covered stents within 72 h
4.9.3. Endoscopic therapy, such as argon plasma coagula- (ideally < 24 h) reduces re-bleeding and improves survival
tion (APC), radiofrequency ablation (RFA), or band in carefully selected patients who met any of the follow-
ligation can be used for PHG and GAVE related bleed- ing criteria: Child–Pugh class C < 14 points or Child–Pugh
ing. (LoE-Low; Recommendation-Strong). class B > 7 with active bleeding at initial endoscopy or
 Hepatology International

HVPG > 16 mmHg at the time of hemorrhage [126]. Patients Consensus statement
fulfilling the criteria for pre-emptive TIPS should be referred
to center with facility to perform TIPS within 72 h if inter- 5. Role of Rescue Therapies in Acute Variceal Bleeding
ventional radiology facilities are not available. TIPS should
be performed in patients having recurrent esophageal
variceal bleed who could not get pre-emptive TIPS. Patients 5.1. Role of balloon tamponade
with uncontrolled variceal bleed should be considered for a 5.1.1. Role of Rescue Therapies in Acute Variceal Bleed-
rescue/salvage TIPS within 24 h, preferably < 8 h. ing
A combination of MELD score and serum lactate levels 5.1.2. Balloon tamponade is effective in immediate and
may provide an improved prediction of survival outcome temporal control of AVB, but is associated with a high
after TIPS. Combined score using MELD > 30 and lac- re-bleeding rate after balloon removal. (LoE-High;
tate > 12 mmol/L is associated with poor survival. Patients Recommendation-Strong)
with acute variceal bleed and ACLF should be cautiously 5.1.3. Balloon tamponade should only be maintained for
evaluated for TIPS based on CTP/MELD score and lactate up to 24 hours to prevent severe adverse effects, such
levels. TIPS may be also considered in patients with HCC as esophageal rupture. (LoE-High; Recommendation-
having acute variceal bleed, if it is technically feasible with Strong)
Child Pugh score. Eight mm TIPS should be preferred over 5.1.4. Guidewire assistance may lead to a higher first-pass
10 mm TIPS for control of bleeding. Embolization of affer- success, proper positioning and reduce the complica-
ent veins feeding the varices and use of beta blockers may be tion rate. (LoE-Moderate; Recommendation-Weak)
utilized if post- TIPS portal pressure gradient of 12 mmHg
is not achieved. TIPS stent of 6 mm may be considered in
patients with small liver / poor hepatic reserve with uncon- 5.2. Role of Self-Expandable Metal Stents (SEMS)
trolled variceal bleed as an emergent salvage pressure. 5.2.1. SEMS is more effective and safer option than bal-
loon tamponade for refractory variceal bleeding. (LoE-
High; Recommendation-Strong)
5.2.2. SEMS is an effective bridge therapy for control
Rescue therapies for gastric varices of refractory variceal bleeding until a more definite
treatment such as TIPS is undertaken. (LoE-High;
BRTO/PARTO should be considered as good alternatives Recommendation-Strong)
for TIPS in patients with GOV2 and IGV1. Patients with 5.2.3. SEMS achieves immediate control of bleeding in
active GV bleed should be considered for endoscopic oblit- up to 90% patients, though is infrequently associated
eration followed by BRTO/PARTO or Salvage TIPS with with adverse events such as stent migration, esopha-
antegrade embolization if feasible (presence of sizable gas- geal ulcerations, and re-bleeding on removal of SEMS.
tro-renal, gastro-spleno-renal shunt) and local expertise is (LoE-Moderate; Recommendation-Strong)
available. BRTO/PARTO should be performed with caution 5.2.4. Esophageal stents are kept in place for up to 5-7
in patients with Child–Pugh score > 9. Suitable patients with days, allowing for a longer bridge than balloon tam-
high-risk GV and ascites may be considered for combina- ponade to definitive therapy. (LoE-Moderate; Recom-
tion of BRTO/PARTO and TIPS, alternatively these patients mendation-Strong)
may be considered for TIPS with antegrade embolization of
all varices (LGV, SGV, PGV). Post BRTO/PARTO patients
should be periodically evaluated with endoscopy to look for 5.3 Role of TIPS (LoE-Moderate; Recommendation-Strong)
any enlargement of esophageal varices. Patients with bleed-
ing from IGV2 can be considered for BRTO/PARTO with
suitable venous anatomy. Alternatively, antegrade emboli- 5.3.1. Pre-emptive TIPS is recommended as the first-line
zation/obliteration of IGV2 may be performed using percu- treatment in high-risk patients with cirrhosis present-
taneous trans-hepatic or trans-splenic access. Radiological ing with AVB, as it significantly improves survival
intervention with TIPS or BRTO can be used as second- outcomes. (LoE-High; Recommendation-Strong)
ary prophylaxis to reduce rebleeding and mortality from 5.3.2. Failure to control variceal bleeding despite com-
GVs. Coil-Assisted Retrograde Transvenous Obliteration bined pharmacological and endoscopic therapy is best
(CARTO) serves as a valuable alternative to BRTO and managed by salvage TIPS to decrease re-bleeding. An
TIPS in patients with GV bleeding. CARTO is a highly 8 mm TIPS should be preferred over 10 mm TIPS.
effective and durable treatment for GV bleeding, with mini- (LoE-Moderate; Recommendation-Strong).
mal complications [127].
Hepatology International

5.3.3. Rescue TIPS should be considered for patients with the procedure selected. (LoE-Moderate; Recommen-
ACLF and variceal bleeding who do not have a con- dation-Strong)
traindication for TIPS. (LoE-Moderate; Recommenda-
tion-Strong)
5.3.4. In patients with cirrhosis presenting with refrac- 5.5. Rescue therapy for gastric variceal bleeding
tory AVB, very early TIPS (emergency TIPS) within
8 hours offers a better survival than TIPS beyond 8
hours. (LoE-Moderate; Recommendation-Strong) 5.5.1. TIPS is an effective rescue therapy to control recur-
5.3.5. Mean arterial pressure, active spurter at endoscopy rent variceal bleeding from gastric or ectopic varices.
and AKI independently predict post-TIPS mortality at (LoE-High; Recommendation-Strong)
6 weeks and one-year. (LoE-Moderate; Recommenda- 5.5.2. Transvenous obliterative procedures like BRTO/
tion-Strong) PARTO are effective therapies to control recurrent or
5.3.6. TIPS may be futile in patients with Child-Pugh difficult to control gastric variceal bleeding in presence
≥ 14 cirrhosis, MELD score >30 and serum lactate of appropriate anatomy. (LoE-High; Recommendation-
>12mmol/L, unless liver transplantation is envisioned Strong)
in the short term. (LoE-High; Recommendation- 5.5.3. BRTO/PARTO and TIPS have comparable efficacy,
Strong) but the former has lower re-bleeding rates and risk for
hepatic encephalopathy. (LoE-High; Recommenda-
tion-Strong)
5.4. Role of surgery 5.5.4. Combining TIPS and embolization of collaterals
can reduce the risk of recurrent gastric variceal bleed-
ing and is a preferred approach. (LoE-Moderate; Rec-
5.4.1. Indication of surgery is persistent variceal bleeding ommendation-Strong)
despite comprehensive medical and endoscopic and
radiological interventions. (LoE-High; Recommenda-
tion-Strong)
5.4.2. Contraindications to surgery are Special topics in AVB

Bacteremia is often present on admission for acute

• Patients with advanced hepatic dysfunction, uncorrect- variceal hemorrhage. Common bacterial infections
able coagulopathy, or hepatic encephalopathy (LoE- include spontaneous bacterial peritonitis, urinary tract
High; Recommendation-Strong) infection, and pneumonia. Infections are associated with
• Child class C categorization, which includes a wide an increased risk of re-bleeding and higher mortality,
spectrum of patients, in itself is not a contraindication likely secondary to a further increase in PHT, further
to emergency surgery. (LoE- Low; Recommendation- splanchnic arteriolar dilatation, and increased coagu-
Weak) lopathy [128, 129]. A complete microbiological work-
up, including blood cultures and diagnostic paracentesis
when appropriate, should be performed. Empiric therapy
5.4.3. Surgery for acute control of variceal bleeding can with a third-generation cephalosporin (e.g., ceftriaxone)
be placed into one of four categories, non-selective should be uniformly instituted because several clinical tri-
portosystemic shunts, selective portosystemic shunts, als have shown improvement in control of bleeding and in
devascularization procedures and liver transplanta- patient outcomes [130]. However, other broad-spectrum
tion. The procedure selected depends on many fac- antibiotics including higher generation of quinolones are
tors, including expertise of available personnel, sever- still acceptable agents.
ity of bleeding, hepatic functional reserve, patency of Prospective cohort studies in which HVPG has been
splanchnic veins, and transplant candidacy. (LoE-Low; measured within 48 h of admission for hemorrhage show
Recommendation-Weak) that levels greater than 20 mmHg are associated with
5.4.4. There is a paucity of controlled data comparing increased re-bleeding and mortality [125, 131–134].
these procedures to one another or to non-operative Another study confirms this HVPG cut-off and shows that
therapies. (LoE-Low; Recommendation-Weak) an index including CTP score and blood pressure at admis-
5.4.5. The major factor determining survival is the status sion has similar prognostic value [135]. Furthermore, a
of the liver disease at the time of surgery rather than vasoactive drug-induced HVPG reduction of less than
10% predicts failure to control bleed. This response may
 Hepatology International

improve by doubling the dose of somatostatin or switching • Patients with diabetes mellitus or CKD
to terlipressin [136]. • Cirrhotic patients with hepatocellular carcinoma

6.2.4. Prevention of hepatic ischemic injury should be

Consensus statement done as follows: (LoE-Low; Recommendation-Weak)
6. Special Topics in Acute Variceal Bleeding
• Resuscitation and adequate correction of hypovolemia,
hypotension, and shock
6.1 Infections in acute variceal bleeding and role of anti-
• Correction of blood loss by PRBC transfusion
biotics. (LoE-High; Recommendation-Strong)
• Rapid control of active bleeding by endoscopy
6.1.1. Chances of developing infection in AVB are high
• Prophylactic antibiotics to guard against sepsis
6.1.2. Gram-negative bacteria, such as E. coli. are com-
mon agents
6.1.3. Placement of various tubes (NG, CVP, ET) predis-
6.2.5. Hepatic ischemic injury could lead to rise in serum
poses to infections and colonizing organisms in stom-
total bilirubin and/or aminotransferases and LDH
ach and skin play a role. Hence, consider changing
within 24 hours and this may adversely affect out-
these tubes regularly. (LoE-Moderate, Recommenda-
comes. (LoE-Low; Recommendation-Weak)
tion-Strong)
6.2.6. Patients should be carefully observed even if hem-
6.1.4. Standard work-up for infection includes CBC, chest
orrhage from varices is controlled. (LoE-Low; Recom-
x-ray, urine and blood culture and serum procalcitonin
mendation-Weak)
estimation. (LoE-High; Recommendation-Strong)
6.2.7. Daily monitoring of ALT, serum bilirubin, creati-
6.1.5. Third-generation cephalosporin is the preferred
nine, LDH, and ALT/LDH ratio should be done when
agent for prevention of infection in AVB and should
ischemic hepatic injury is suspected. (LoE-Low; Rec-
be given intravenous
ommendation-Weak)
6.2.8. No definite therapy has been found for the treatment
6.2. Prevention, assessment, and management of
of ischemic hepatic injury, but N-acetyl cysteine may
hepatic ischemia. (LoE-High; Recommendation-
be helpful. (LoE-Low; Recommendation-Weak)
Strong)
6.2.9.There is a need for prospective studies to further inves-
6.2.1. Ischemic hepatic injury can occur in up to 10%
tigate the prevalence, severity, and treatment of hepatic
cirrhosis patients with AVB
ischemia in cirrhosis patients with variceal bleeding.
6.2.2. Hepatic ischemic injury should be anticipated and
6.3. Acute variceal bleeding in patients with liver failure
prevented in high-risk groups of patients with
6.3.1. Patients with cirrhosis and liver failure have
bleeding varices. (LoE-Low; Recommendation-
high propensity for bleeding from gastro-esophageal
Weak)
varices. (LoE-Low; Recommendation-Weak)
6.2.3. The following groups of patients have high risk
6.3.2. Endoscopic procedures to control variceal bleed-
of hepatic ischemic injury:
ing should be done with extra caution with or without
  enously for 5 days. endotracheal intubation. (LoE-Low; Recommendation-
Strong)
6.3.3. Coagulopathy should be corrected with FFP and
• Patients with severe hematemesis and melena platelets. (LoE-Low; Recommendation-Weak)
• Bleeding leading to significant hypotension and/or shoc 6.3.4. Role of factor rVIIa in this setting needs evaluation.
• Recurrent bouts of bleeding at home, during transfer, (LoE-Low; Recommendation-Weak)
at the casualty department, in the ward, and before or
during emergency endoscopy
• Re-bleeding in a known patient with history of variceal Diagnosis and treatment of acute ectopic
bleeding variceal bleeding
• Significant drop of hemoglobin and hematocrit values
• Patients with obstruction to hepatic blood flow (portal Ectopic varices comprise large portosystemic venous col-
vein thrombosis, Budd-Chiari syndrome) laterals located anywhere other than the gastro-esophageal
• Patients with decompensated cirrhosis region [137]. No large series or RCTs address this sub-
• Elderly patients ject, and therefore their management is based on available
Hepatology International

expertise and facilities and may require a multidisciplinary 7.1. Bleeding ectopic varices are a rare cause of variceal
team approach. Ectopic varices are common findings during bleeding and are common in non-cirrhotic patients.
endoscopy in portal hypertensive patients. Bleeding ectopic (LoE-Moderate; Recommendation-Weak)
varices are a rare cause of variceal bleeding and accounts for 7.2. Site (LoE-Moderate; Recommendation-Weak)
only 1–5% of all variceal bleeding [138]. 7.2.1. Ectopic varices occur in anorectum, antrum (IGV2),
Ectopic varices occur in anorectum, antrum (IGV2), and and duodenum, small intestine, colon, and peristomal
duodenum, small intestine, colon, and peristomal. Ectopic 7.2.2. Ectopic variceal bleeding may be from varices
variceal bleeding may be from varices located in the fol- located in the following sites: duodenum, choledochal,
lowing sites: duodenum, choledochus, omentum, stoma, and omentum, stoma, and rectum
rectum. Bleeding is more frequent in peristomal varices. 7.2.3. Bleeding is more frequent in peristomal varices
Ectopic varices develop secondary to portal hyperten- 7.3. Endoscopy can diagnose most of the cases, but in inac-
sion, surgical procedures, anomalies in venous outflow, cessible sites, RBC scan helps to identify the site of
or abdominal vascular thrombosis and may be familiar in bleed, which is to be confirmed by CT angiography.
origin. Bleeding ectopic varices may present with anemia, (LoE-Moderate; Recommendation-Weak)
shock, hematemesis, melena, or hematochezia and should be 7.4. One of the following endoscopic findings constitutes
considered in patients with PHT and gastrointestinal bleed- acute ectopic variceal bleeding: (LoE-Moderate; Rec-
ing or anemia of obscure origin. Ectopic varices may be ommendation-Weak)
discovered during pan-endoscopy, enteroscopy, endoscopic
ultrasound, wireless capsule endoscopy, diagnostic angiog-
raphy, multi slice helical computed tomography, magnetic • Direct visualization of blood issuing from varix— usu-
resonance angiography, color Doppler-flow imaging, lapa- ally spurting
rotomy, laparoscopy, and occasionally during autopsy [139]. • Presence of a sign of recent bleed: white nipple sign or
Patients with suspected ectopic variceal bleeding need overlying clot
immediate assessment, resuscitation, hemodynamic stabili-
zation, and referral to specialist centers. Endoscopy proce-
dure able to diagnose most of the cases, but in inaccessible 7.5. Pharmacotherapy and endotherapy should be the first
sites, red blood cell (RBC) evaluation can identify the site of line of therapy if a bleeding ectopic varix is accessible.
bleed and could be confirmed by angiography or CT angiog- In inaccessible cases, TIPS or PTVE should be done
raphy. One of the following endoscopic findings constitutes with patent portal vein. (LoE-Low; Recommendation-
acute ectopic variceal bleeding: direct visualization of blood Weak)
issuing from varix—usually spurting; presence of a sign of 7.6. Duodenal variceal bleeding inaccessible by endoscopy
recent bleed, white nipple sign, or overlying clot. can also have an option of BRTO (if vascular anatomy
Management of ectopic varices involves medical, endo- permits) and EUS guided embolization. (LoE-Low;
scopic, interventional radiological, and surgical modalities Recommendation-Weak)
depending on the patient condition, site of varices, avail-
able expertise and patient’s subsequent management plan.
Pharmacotherapy and endotherapy should be the first line of Acute variceal bleed in non‑cirrhotic portal
therapy if a bleeding ectopic varix is accessible, but in inac- hypertension (NCPH)
cessible cases, TIPS or percutaneous transhepatic varices
embolization (PTVE) should be done in patients with patent Variceal bleeding is a common and life-threatening compli-
portal vein in cirrhosis and NCPH. Duodenal variceal bleed- cation of PHT due to NCPH. There is paucity of data on the
ing inaccessible by endoscopy can also have an option of management of AVB in NCPH; however, the principles and
BRTO if vascular anatomy permits [137, 139, 140]. modes of management remain the same as those for patients
with cirrhosis. Blood transfusion, intravenous fluids, and
standard ICU care are provided [1, 141]. Bacterial infections
are more common in patients with cirrhosis having variceal
Consensus statement bleeding (35–66%) than in non-cirrhotic patients (5–7%)
7. Diagnosis and Treatment of Acute Ectopic Variceal [142]. It has been shown that infected cirrhotic patients have
Bleeding a higher rate of variceal re-bleeding (43%) than non-infected
patients (10%) [128]. In patients with cirrhosis and variceal
 Hepatology International

bleeding, prophylactic antibiotics reduce variceal re-bleed- 8.7. Antibiotics are not recommended in AVB in NCPF,
ing and improve survival [130]. In NCPH, however, there is unless absolute neutrophil count is <1,000 m ­ m-3.
no study on the use of prophylactic antibiotics. (LoE-Low; Recommendation-Weak)
Vasoactive drugs, such as somatostatin, octreotide, or 8.8. Rescue therapies remain the same as in cirrhosis
terlipressin have been used in the treatment of AVB while patients. (LoE-Low; Recommendation-Weak)
endoscopic therapy is being arranged. The vasoactive drugs 8.9. Radiological interventions as rescue therapy in NCPH:
lead to reduction in portal pressure, which is associated though no randomized control trials have been con-
with a better control of variceal bleeding [143–145]. How- ducted to investigate the potential of these techniques,
ever, there are no data on the efficacy of vasoactive drugs in case reports and series suggest their efficacy for con-
patients with NCPH with AVB. trolling variceal bleeding. (LoE-Low; Recommenda-
Endoscopic sclerotherapy and band ligation are effective tion-Weak)
in 80–90% of patients in controlling acute bleeding from 8.10. Factors influencing choice of radiological procedure
EVs and preventing re-bleeding. At present, band ligation include etiological and anatomical considerations,
is preferred owing to lower complication rates. Combina- clinical status of the patient and available expertise
tion treatment with drugs plus endoscopic therapy is more and affordability. (LoE-Low; Recommendation-Weak)
effective than endoscopic therapy or drug therapy alone in 8.11. Though data is limited, it is likely that complications of
controlling acute bleeding and preventing re-bleeding for TIPS would be uncommon in NCPF patients due to the
5 days while there is no difference in mortality [1, 140]. relatively preserved liver functions in them. (LoE-Low;
There is, however, paucity of data in patients with NCPH. Recommendation-Weak)
Failure of endoscopic therapy is defined, as further variceal 8.12. Patients with failed first-line therapy for variceal bleed-
bleeding after two endoscopic treatments during a single ing and radiological interventional approaches should
hospital admission for acute bleeding. The current therapies be considered for surgery. (LoE-Moderate; Recom-
fail to control bleeding or prevent early re-bleeding in 8–12% mendation-Weak)
of patients, who should be treated by alternative modes of 8.13. Portal decompressive procedures are better than non-
treatment, like surgery or TIPS. shunt procedures. Non-shunt procedures are preferred
in patients who do not have suitable veins. (LoE-Low;
Recommendation-Weak)
Consensus statement
8. Acute Variceal Bleed in NCPH
Pediatric perspectives of AVB

Evidence-based approaches to the management of adults

8.1. Absence of ascites, jaundice, and hepatic encepha-
with AVB exist and have been comprehensively reviewed
lopathy, and presence of large splenomegaly are the
[1]. Similar evidence-based approaches for the management
clinical clues in differentiating NCPH from cirrhotic
of AVB in children do not exist and as such most interna-
portal hypertension (CPHT). (LoE-Moderate; Recom-
tional meetings on PHT have not focused on this problem
mendation-Strong)
in children [148]. Approaches to the management of AVB
8.2. Natural history of acute variceal bleeding in NCPH
in children are few. Therefore, pediatricians typically have
has been well studied and shows low mortality. (LoE-
difficulty in deciding how to manage this important clini-
Moderate; Recommendation-Strong)
cal problem in children. The statements presented here are
8.3. Definitions and time frames for acute variceal bleeding
mostly expert opinion with evidence being extrapolated from
as for cirrhosis can be adapted for NCPH patients as
the studies done in adults.
well. (LoE-Low; Recommendation-Weak)
8.4. First-line treatment options are essentially the same
as in cirrhosis patients. (LoE-Low; Recommendation-
Consensus statement
Weak)
8.5. Gastric varices are more common in NCPH and may 9. Pediatric Perspectives of Acute Variceal Bleeding
be refractory to obturation by tissue adhesives. (LoE-
Low; Recommendation-Weak)
8.6. Coagulopathy is generally not a feature of NCPH, and 9.1. The pediatric age-group is defined as age up to 18
so, correction is not required. (LoE-Low; Recommen- years. (LoE-Low; Recommendation-Weak)
dation-Weak)
Hepatology International

9.2. The majority of upper gastro-intestinal bleed in chil- therapeutic strategies tailored to clinical practice in the
dren is variceal in origin. (LoE-Moderate; Recommen- Asia–Pacific region.
dation-Strong)
9.3. The etiology of acute variceal bleeding in children 1. Refined Definitions and Risk Stratification
varies in different geographical regions: in the West,   The updated guidelines retain the core definitions of
cirrhosis is more common while in the East, EHPVO AVB and re-bleeding but now integrate home-to-door
is more common. (LoE-Low; Recommendation-Strong concept, management during patient transfer and in ER,
9.4. Diagnosis and management are broadly similar to that validated risk prediction models, including the APASL
in adults [1], with some exceptions: Bleed Severity Score, HVPG-guided risk assessment,
and AI-assisted prognostication. This ensures more pre-
cise risk stratification and individualized management.
• Restrictive transfusion threshold (within the range of 2. Expanded Role of Early and Rescue Therapies
7–8 g/dL) in pediatric variceal bleeding, as long as the   While the 2011 guidelines recommended balloon tam-
child is hemodynamically stable, with slightly higher ponade and TIPS as rescue therapies, the 2025 guide-
targets considered for specific cases, such as those with lines redefine when and how to escalate therapy. The
cyanotic heart disease or significant hypoxia [146]. (LoE- updated recommendations now incorporate:
Low; Recommendation-Weak)
• Endoscopy performed within 12 to 24 h yields similar
outcomes compared to more urgent procedures [147]. • Use of self-expanded metal stents (SEMS) for control of
(LoE-Moderate; Recommendation-Strong) AVB
• Early-TIPS strategy, particularly for patients with
HVPG > 20 mmHg.
9.5. Dosage and safety profile of octreotide in children has • EUS-guided glue and thrombin injection as alternatives
been established, however, for terlipressin or soma- for glue injection in acute gastric variceal bleeding.
tostatin, the dose and safety need to be established in • BRTO and CARTO/PARTO as key interventions in man-
children. (LoE-Low; Recommendation-Weak) aging gastric varices, especially in patients with gastro-
9.6. Choice of endoscopic procedure: (LoE-Low; Recom- renal shunts.
mendation-Weak)

3. Personalized Approach to Antibiotic Therapy and Phar-

• Band ligation is preferred over EST for acute variceal macologic Management
bleeding   The updated guidelines emphasize region-specific
• EST is technically more feasible in younger children and antibiotic choices, moving beyond a one-size-fits-all
those with smaller varices. ceftriaxone approach. The selection of terlipressin,
octreotide, or somatostatin has also been refined based
on timing, severity, and patient comorbidities.
9.7. Rescue therapies 4. Standardization of Endoscopic Therapy and Anesthesia
9.7.1. Radiological: though no randomized control tri- Protocols
als have been conducted to investigate the potential   EVL remains the gold standard, but the new guide-
of radiological techniques in children, case reports lines address anesthesia selection and peri-procedural
and case series suggest their efficacy for controlling risk assessment, ensuring greater safety in cirrhosis
variceal bleeding. (LoE-Low; Recommendation-Weak) patients undergoing endoscopic interventions.
9.7.2. Radiological: though no randomized control tri- 5. Greater Emphasis on Clinical Practice Implementation
als have been conducted to investigate the potential   EVL remains the gold standard, but the new guide-
of radiological techniques in children, case reports lines address anesthesia selection and peri-procedural
and case series suggest their efficacy for controlling risk assessment, ensuring greater safety in cirrhosis
variceal bleeding. (LoE-Low; Recommendation-Weak) patients undergoing endoscopic interventions.

What’s New in the APASL AVB Guidelines? The updated APASL guidelines provide a more dynamic,
The APASL 2025 guidelines on AVB represent a major evidence-based, and practical approach to managing AVB,
advancement over the 2011 recommendations, incorporat- ensuring earlier intervention, better risk stratification, and
ing the latest evidence, refined definitions, and improved improved patient outcomes. These updates align with global
 Hepatology International

standards while addressing the unique epidemiology and References

healthcare challenges of the Asia–Pacific region.
1. de Franchis R, Bosch J, GarciaTsao G, Reiberger T, Ripoll C;
Baveno VII Faculty.
2. Baveno VII – Renewing consensus in portal hyperten-
Future research topics sion: personalized care for portal hypertension. J Hepa-
tol. 2022;76(4):959–974. 2.Stanley A, Laine LB. Management
Advancing the management of AVB will require multicen- of acute upper gastrointestinal bleeding. BMJ. 2019;364:1536.
tric studies to fill the current knowledge gaps. The prior- 3. Kaplan D, Ripoll C, Thiele M, et al. AASLD Practice Guidance
on risk stratification and management of portal hypertension and
ity areas include choice and efficacy of vasoactive drugs, varices in cirrhosis. Hepatology. 2024;79(5):1189–1211
portal pressure gradient measurement by EUS during AVB 4. Sarin SK, Sollano JD, Chawla YK, Amarapurkar D, Hamid
compared with HVPG compared and role of TIPS as a pre- S, Hashizume M, et al. Consensus on extrahepatic portal vein
emptive or a rescue therapy. obstruction. Liver Int. 2006;26(5):512–519
5. Sarin S, Kumar A, Chawla Y, et al. Noncirrhotic portal fibrosis/
idiopathic portal hypertension: APASL recommendations for
diagnosis and treatment. Hepatol Int. 2007;1:398–413
Conclusions 6. Sarin SK, Kumar A, Angus PW, Baijal SS, Chawla YK, Dhi-
man RK, et al. Primary prophylaxis of gastroesophageal variceal
bleeding: consensus recommendations of the Asian Pacific Asso-
The new APASL consensus guidelines on AVB assimilate ciation for the Study of the Liver. Hepatol Int. 2008;1(4):398–413
the latest evidence in the management of AVB and will con- 7. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R,
tribute to a better care of patients suffering from this condi- Brozek J, et al. GRADE guidelines: 3. Rating the quality of evi-
tion. The guidelines aimed to improve the quality of care dence. J Clin Epidemiol 2011 04;64(4):401–6.
8. Barbu L, Margaritescu N, Surlin M. Diagnosis and treat-
and reduce the morbidity and mortality following an AVB. ment algorithms of acute variceal bleeding. Curr Health Sci.
These guidelines have considered all available data and 2017;43(3):191–200
guidelines in the field to finalize the consensus statements, 9. de Franchis R. Diagnosis and management of acute variceal
which were achieved with unanimity. These AVB guidelines bleeding. Clin Liver Dis. 2010;14(4):1–22.c
10. de Franchis. Somatostatin, somatostatin analogues and other vas-
are intended to provide the physicians and multi-disciplinary oactive drugs in the treatment of bleeding oesophageal varices.
teams evidence-based clear roadmaps for patient stratifica- Dig Liver Dis. 2000;36 (suppl.1): S93-S100)
tion, strategizing timely interventions with judicious use of 11. de Franchis R. Developing consensus in portal hypertension. J
resources for the best possible outcomes of AVB patients. Hepatol. 1996;25:390–394
12. de Franchis R (editor). Portal Hypertension II. Proceedings of
The experts have also highlighted the issues which could the Second Baveno International Consensus Workshop on Defini-
not be addressed due to lack of adequate evidence. The con- tions, Methodology and Therapeutic Strategies. Oxford: Black-
sensus definitions are intended to be used in trials and other well Science; 1996.
studies on AVB, as well as in clinical practice. It is desirable 13. de Franchis R. Updating consensus in portal hypertension: report
of the Baveno III consensus workshop on definitions, methodol-
that future studies be reported using these definitions as part ogy and therapeutic strategies in portal hypertension. J Hepatol.
of the evaluation. This should result in some measure of 2000;33:846–852
standardization and increased ease of interpretation among 14. de Franchis R (editor). Portal Hypertension III. Proceedings of
different studies. Equally important, if there are uniformly the IIIrd Baveno International Consensus Workshop on Defini-
tions, Methodology and Therapeutic Strategies. Oxford: Black-
defined endpoints, meta-analyses will be based on more well Science; 2001.
homogeneous studies, which is an essential prerequisite of 15. de Franchis R. Evolving consensus in portal hypertension.
this methodology. It is hoped that these guidelines will move Report of the Baveno IV consensus workshop on methodol-
the field of management of AVB to a much higher level. ogy of diagnosis and therapy in portal hypertension. J Hepatol.
2005;43(1):167–176
Supplementary Information The online version contains supplemen- 16. Sarin SK, Kumar A, Angus PW, et al. Diagnosis and management
tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 12072-0​ 25-1​ 0894-4. of acute variceal bleeding: Asian Pacific Association for Study
of the Liver recommendations. Hepatol Int. 2011;5(2):607–624
Acknowledgements We thank to Prof. Diana A. Payawal, Fatima Uni- 17. Cales P, Masliah C, Bernard B, Garnier PP, Silvain C, Szos-
versity Medical Center, for her contributions to this guideline during takTalbodec N, et al. Early administration of vapreotide for
revision and finalization variceal bleeding in patients with cirrhosis. N Engl J Med.
2001;344(1):23–28
Data availability Not Applicable. 18. Kollef M, O’Brien J, Zuckerman G, et al. BLEED: A classifica-
tion to predict outcome in patients with acute upper and lower
gastrointestinal hemorrhage. Crit CareMed. 1997;25:1101–1012
19. Eslam M, Sanyal A, George J, et al. MAFLD: a consensus-driven
proposed nomenclature for metabolic associated fatty liver dis-
ease. Gastroenterol. 2020;158:1999=2014.
Hepatology International

20. Sutton F. Upper gastrointestinal bleeding in patients with esoph- review and meta-analysis. World J Gastroenterol. 2020. https://​
ageal varices: what is the most common source? Am J Med. doi.​org/​10.​3748/​wjg.​v26.​i18.​2247
1987;32(3):273–275 40. Rotondano G. Epidemiology and diagnosis of acute nonvariceal
21. Haj M, Rockey D. Predictors of clinical outcomes in cirrhosis upper gastrointestinal bleeding. Gastroenterol Clin North Am.
patients. Curr Opin Gastroenterol. 2018;34(4):266–271 2014;43:643–663
22. Lemmer P, Pospiech J, Canbay A. Liver failure—future chal- 41. Hreinsson J, Kalaitzakis E, Gudmundsson S, Björnsson E.
lenges and remaining questions. Ann Transl Med. 2021;9(8):734 Upper gastrointestinal bleeding: incidence, etiology and out-
23. Garcia-Tsao G, Abraldes J, Berzigotti A, Bosch J. Portal hyper- comes in a population-based setting. Scand J Gastroenterol.
tensive bleeding in cirrhosis: risk stratification, diagnosis, and 2013;48:439–447
management: 2016 practice guidance by the American Associa- 42. Wuerth B, Rockey D. Changing epidemiology of upper gastro-
tion for the study of liver diseases. Hepatol. 2017;65(1):310–335 intestinal hemorrhage in the last decade: a nationwide analysis.
24. Berzigotti A, Seijo S, Arena U, et al. Elastography, spleen size, and Dig Dis Sci. 2014;63(5):1286–1293
platelet count identify portal hypertension in patients with compen- 43. Lanas A, Dumonceau J, Hunt R, et al. Non-variceal upper gas-
sated cirrhosis. Gastroenterol. 2013;144(1):102-111.e101 trointestinal bleeding. Nat Rev Dis Primers. 2018;4: 18020
25. Kamboj A, Hoversten P, Leggett C. Upper gastrointesti- 44. Barkun A, Leontiadis G. Systematic review of the symptom bur-
nal bleeding: etiologies and management. Mayo Clin Proc. den, quality of life impairment and costs associated with peptic
2019;94(4):697–703 ulcer disease. Am J Med. 2010;123:358–366
26. Simon-Talero M, Roccarina D, Martinez J, et al. Associa- 45. Forrest J, Finlayson N, Shearman D. Endoscopy in gastrointes-
tion between portosystemic shunts and increased complica- tinal bleeding. Lancet. 1974;2:394–397
tions and mortality in patients with cirrhosis. Gastroenterol. 46. Heldwein W, Schreiner J, Pedrazzoli J, Lehnert P. Is the Forrest
2018;154(6):1694-1705.e1694 classification a useful tool for planning endoscopic therapy of
27. Vidal-Gonzales J, Martinez J, Mulay A, et al. Evolution of spon- bleeding peptic ulcers? Endoscopy. 1989;21:258–262
taneous portosystemic shunts over time and following aetiologi- 47. Lau J, et al. The evolution of stigmata of hemorrhage in bleed-
cal intervention in patients with cirrhosis. JHEP Rep. 2023;30(2): ing peptic ulcers: a sequential endoscopic study. Endoscopy.
100977 1998;30:513–518
28. Praktiknjo M, Simon-Talero M, Romer J, et al. Total area of 48. Almadi M, Lu Y, Alali A, Barkun A. Peptic ulcer disease. Lancet.
spontaneous portosystemic shunts independently predicts hepatic 2024;404(10447):68–81
encephalopathy and mortality in liver cirrhosis. J Hepatol. 49. Aubé C, Bazeries P, Lebigot J, Cartier V, Boursier J. Liver fibro-
2020;72(6):1140–1150 sis, cirrhosis, and cirrhosis-related nodules: imaging diagnosis
29. Zardi E, Uwechie V, Caccavo D, et al. Portosystemic shunts in and surveillance. Diagn Interv Imaging. 2017;98:455–468
a large cohort of patients with liver cirrhosis: detection rate and 50. Jung JH, Jo JH, Kim SE, Bang CS, Seo SI, Park CH, et al.
clinical relevance. J Gastroenterol. 2009;44(1):76–83 Minimal and maximal extent of band ligation for acute variceal
30. Zhou H, Chen T, Zhang X, et al. Diameters of left gastric vein bleeding during the first endoscopic session. Gut Liver.
and its originating vein on magnetic resonance imaging in liver 2022;16(1):101–110
cirrhosis patients with hepatitis B: association with endoscopic 51. Gralnek I, Camus Duboc M, Garcia-Pagan J, et al. Endoscopic
grades of esophageal varices. Hepatol Res. 2014;44:E110-117 diagnosis and management of esophagogastric variceal hem-
31. Iwakiri Y. Pathophysiology of portal hypertension. Clin Liver orrhage: European Society of Gastrointestinal Endoscopy
Dis. 2014;18:281–291 (ESGE) guideline. Endoscopy. 2022;54(11):1094–1120
32. Chen X, Chen T, Zhang X, et al. Platelet count combined with 52. Sarin S, D L, Saxena S, Murthy N, Makwana U. Prevalence,
right liver volume and spleen volume measured by magnetic res- classification and natural history of gastric varices: a long-term
onance imaging for identifying cirrhosis and esophageal varices. follow-up study in 568 portal hypertension patients. Hepatol-
World J Gastroenterol. 2015;21(35):10184–10191 ogy 1992;16(6):1343–1349.
33. Klasen J, Lanzman R, Wittsack H, et al. Diffusion-weighted imag- 53. Kupfer Y, Cappell M, Tessler S. Acute gastrointestinal bleed-
ing (DWI) of the spleen in patients with liver cirrhosis and portal ing in the intensive care unit. The intensivist’s perspective.
hypertension. Magn Reson Imaging. 2013;31(7):1092–1096 Gastroenterol Clin North Am. 2000;29(2):275–307
34. Razek A, Massoud S, Azziz M, El-Bendary M, Zalata K, Mot- 54. Wong YJ, Buckholz A, Sim A, et al. Nonalcohol-related cir-
awea E. Prediction of esophageal varices in cirrhotic patients rhosis leads to higher 6-week mortality after acute variceal
with apparent diffusion coefficient of the spleen. Abdom Imag- bleeding than alcohol-related cirrhosis. Clin Gastroenterol
ing. 2015;40(6):1465–1469 Hepatol. 2024. 54.
35. Ronot M, Lambert S, Elkrief L, et al. Assessment of portal hyper- 55. Villagrasa A, Hernández-Gea V, Bataller R, et al. Alcohol-
tension and high-risk oesophageal varices with liver and spleen related liver disease phenotype impacts survival after an acute
three-dimensional multifrequency MR elastography in liver cir- variceal bleeding episode. Liver Int. 2023;43(7):1548–1557
rhosis. Eur Radiol. 2014;24(6):1394–1402 56. Elhendawy M, Eldesouky A, Soliman S, Mansour L, AbdEl-
36. Shin S, Lee J, Yu M, et al. Prediction of esophageal varices in salam S, Hawash N. AIMS65 and PALBI scores as predictors of
patients with cirrhosis: usefulness of three-dimensional MR six months’ mortality in cirrhotic patients with acute variceal
elastography with echo-planar imaging technique. Radiology. bleeding. Open Biomark J. 2022;12: e187531832207040
2014;272(1):143–153 57. Schuppan D, Afdhal N. Liver cir rhosis. Lancet.
37. Palaniyappan N, Cox E, Bradley C, et al. Non-invasive assess- 2008;371:838–851
ment of portal hypertension using quantitative magnetic reso- 58. Berg T. Diagnostics for elevated liver values. Gastroenterologe.
nance imaging. J Hepatol. 2016;65(6):1131–1139 2009;2009(4):557–572
38. Pallio S, Melita G, Shahini E, et al. Diagnosis and management 59. Ginès P, Krag A, Abraldes J, Solà E, Fabrellas N, Kamath P.
of esophagogastric varices. Diagnostics. 2023;13: 1031 Liver cirrhosis. Lancet. 2021;398(10308):1359–1376
39. Li Y, Li L, Weng H, Liebe R, Ding H. Computed tomography vs 60. Shung D, Au B, Taylor R, et al. Validation of a machine
liver stiffness measurement and magnetic resonance imaging in learning model that outperforms clinical risk scoring sys-
evaluating esophageal varices in cirrhotic patients: a systematic tems for upper gastrointestinal bleeding. Gastroenterol.
2019;158(1):160–167
 Hepatology International

61. Levi R, Carli F, Arévalo A, et al. Artificial intelligence-based 79. Chan W, Han J, Xiao L. Efficacy of vasopressin/terlipressin
prediction of transfusion in the intensive care unit in patients and somatostatin /octreotide for the prevention of early variceal
with gastrointestinal bleeding. BMJ Health Care Inform. rebleeding after the initial control of bleeding: a systematic
2021;28(1): e100245 review and meta-analysis. Hepatol Int. 2015;9:120–129
62. Alshamsi F, Jaeschke R, Baw B, Alhazzani W. Prophylactic 80. Biswas S, Lo GH, Mehta S, Elhence A, Wong YJ, Vaishnav
endotracheal intubation in patients with upper gastrointestinal M, et al. Abbreviated duration of vasoactive agents has similar
bleeding undergoing endoscopy: a systematic review and meta- outcomes as standard duration of therapy in patients with liver
analysis. Saudi J Med Sci. 2017;5(3):201–209 cirrhosis and variceal bleeding: an individual patient data meta-
63. Kravetz D, Bosch J, Arderiu M, et al. Hemodynamic effects analysis. Dig Dis Sci. 2025;70(3):1201–1214
of blood volume restitution following a hemorrhage in rats 81. Arora V, Choudhary S, Maiwall R, et al. Low-dose continu-
with portal hypertension due to cirrhosis of the liver: influ- ous terlipressin infusion is effective and safer than intravenous
ence of the extent of portal-systemic shunting. Hepatology. bolus injections in reducing portal pressure and control of acute
1989;9:808–814 variceal bleeding. Hepatol Int. 2023;17(1):131–138
64. Kravetz D, Sikuler E, Groszmann R. Splanchnic and systemic 82. Lin L, Cui B, Deng Y, Jiang X, Liu W, Sun C. The efficacy of
hemodynamics in portal hypertensive rats during hemorrhage and proton pump inhibitor in cirrhotics with variceal bleeding: a sys-
blood volume restitution. Gastroenterolol. 1986;90:1232–1240 temic review and meta-analysis. Digestion. 2021;102(2):117–127
65. Biswas S, Vaishnav M, Pathak P, et al. Effect of thrombocy- 83. Hidaka H, Nakazawa T, Wang G, et al. Long-term administra-
topenia and platelet transfusion on outcomes of acute variceal tion of PPI reduces treatment failures after esophageal variceal
bleeding in patients with chronic liver disease. World J Hepatol. band ligation: a randomized,controlled trial. J Gastroenterol.
2022;14(7):1421–1437 2012;47:118–126
66. Mohanty A, Kapuria D, Canakis A, et al. Fresh frozen plasma 84. Zhu J, Qi X, Yu H, Su C, Guo X. Acid suppression in patients
transfusion in acute variceal haemorrhage: results from a multi- treated with endoscopic therapy for the management of gastroe-
centre cohort study. Liver Int. 2021;41(8):1901–1908 sophageal varices: a systematic review and meta-analysis. Expert
67. Rout G, Shalimar, Gunjan D, et al. Thromboelastography-guided Rev Gastroenterol Hepatol. 2018;12(6):617–624
blood product transfusion in cirrhosis patients with variceal 85. Shaheen N, Stuart E, Schmitz S, et al. Pantoprazole reduces the
bleeding: A randomized controlled trial. J Clin Gastroenterol. size of postbanding ulcers after variceal band ligation: a rand-
2020;54(3):255–262. omized, controlled trial. Hepatology. 2005;41(3):588–594
68. Agha O, Alsayid M, Reynolds J. Nasoenteric tube placement in 86. Lo W, Wilby K, Ensom M. Use of proton pump inhibitors in the
patients with esophageal varices: a review of current evidence management of gastroesophageal varices: a systematic review.
and society guideline. Avicenna J Med. 2023;13(4):193–198 Ann Pharmacother. 2015;49(2):207–219
69. Mallet M, Rudler M, Thabut D. Variceal bleeding in cirrhotic 87. Lashen SA, Shamseya MM, Shamseya AM, Hablass FA. Effi-
patients. Gastroenterol Rep. 2017. https://d​ oi.o​ rg/1​ 0.1​ 093/g​ astro/​ cacy of vonoprazan vs. pantoprazole or non-acid suppression
gox024 in prevention of post-variceal ligation ulcer bleeding in portal
70. Fernández J, Navasa M, Gómez J, et al. Bacterial infections in hypertension: a multiarm randomized controlled trial. J Clin
cirrhosis: epidemiological changes with invasive procedures and Exp Hepatol. 2023;13(6):962–971
norfloxacin prophylaxis. Hepatology. 2002;35(1):140–148 88. Lashen S, Shamseya M, Shamseya A, Hablass F. Efficacy of
71. Runyon B. Spontaneous bacterial peritonitis in adults: treatment vonoprazan vs. pantoprazole or non-acid suppression in pre-
and prophylaxis, UpToDate. 2023; https://​www.​uptod​ate.​com/​ vention of post-variceal ligation ulcer bleeding in portal hyper-
conte​nts/​spont​aneous-​bacte​rial-​perit​onitis-​in-​adults-​treat​ment-​ tension: a multi-arm randomized controlled trial. J Clin Exp
and-​proph​ylaxis?​secti​onName=​Choos​ing+a+​regim​en&​topic​ Hepatol. 2023;13:962–971
Ref=1​ 254&a​ nchor=H ​ 80470​ 555& source=see_link#H80470555. 89. Kumar M, Venishetty S, Jindal A, et al. Tranexamic acid in
72. Wells M, Chande N, Adams P, et al. Meta-analysis: vasoactive upper gastrointestinal bleed in patients with cirrhosis: a rand-
medications for the management of acute variceal bleeds. Ali- omized controlled trial. Hepatology. 2024;80(2):376–388
ment Pharmacol Ther. 2012;35:1267–1278 90. Veitch AM, Vanbiervliet G, Gershlick AH, et al. Endoscopy
73. de Franchis R. Expanding consensus in portal hypertension in patients on antiplatelet or anticoagulant therapy, including
Report of the Baveno VI Consensus Workshop: Stratifying risk direct oral anticoagulants: British Society of Gastroenterology
and individualizing care for portal hypertension. J Hepatol. (BSG) and European Society of Gastrointestinal Endoscopy
2015;63:743–752 (ESGE) guidelines. Endoscopy. 2021;53(9):947–969
74. Shah H, Mumtaz K, Jafri W, et al. The role of octreotide in acute 91. Hou M, Lin H, Liu T, et al. Antibiotic prophylaxis after endo-
variceal bleeding; a prospective, randomized study. Med Coll scopic therapy prevents rebleeding in acute variceal hemor-
Abbottabad. 2005;17(1):10–14 rhage: a randomized trial. Hepatology. 2004;39:746–753
75. Huaringa-Marcelo J, Huaman M, Condorena A, et al. Vaso- 92. Walter A, Marika R, O P, Lucile M, Trépo E, Angèle R. Com-
active agents for the management of acute variceal bleeding: bination of model for end‐stage liver disease and lactate pre-
a systematic review and meta-analysis. Gastrointest Liver Dis. dicts death in patients treated with salvage transjugular intra-
2021;30(1):110–121 hepatic portosystemic shunt for refractory variceal bleeding.
76. Azam Z, Hamid S, Jafri W. Short course adjuvant terlipressin Hepatol. 2021;74(4):2085–2101.
in acute variceal bleeding: a randomized double blind dummy 93. Magifira N, Rozzaqi A, Jasiwrwan C, Suhartono R. Surgical
controlled trial. J Hepatol. 2012;56(4):819–824 shunting versus TIPS for recurrent variceal bleeding in liver
77. Sudhagar R, Sheik A, Sarath S, et al. Comparison of 2 days ver- cirrhosis: An evidence-based case report. The Indonesion J
sus 5 days of octreotide infusion along with endoscopic therapy Gastroenterolol, hepatol digest. 2023;24:264–269
in prevention of early rebleed from esophageal varices: a rand- 94. Rodrigues S, Cárdenas A, Escorsell A, Bosch J. Balloon tam-
omized clinical study. Eur J Gastroenterol. 2015;27:386–392 ponade and esophageal stenting for esophageal variceal bleed-
78. Abid S, Jafri W, Saeed H, et al. Terlipressin vs octreotide in ing in cirrhosis: a systematic review and meta-analysis. Semin
bleeding esophageal varices as adjuvant therapy with endoscopic Liver Dis. 2019;39(2):178–194
band ligation: a randomized double blind placebo controlled trial.
Am J Gatroenterolol. 2009;104:617–623
Hepatology International

95. Peng M, Bai Z, Zhou D, et al. Timing of endoscopy in patients (CHESS1905): a nationwide cohort study. Front Med. 2022;9:
with cirrhosis and acute variceal bleeding: a single-center ret- 872881
rospective study. BMC Gastroenterol. 2023;23:219 114. Lu Z, Sun X, Zhang W, et al. Second urgent endoscopy within
96. Yan X, Leng Z, Xu Q, Zhang Z, Xu M, Li J. The influences 48-hour benefits cirrhosis patients with acute esophageal variceal
of timing of urgent endoscopy in patients with acute variceal bleeding. Medicine. 2020;99(11): e19485
bleeding: a cohort study. BMC Gastroenterol. 2022;22(1):506 115. Jamwal K, Kumar M, Maiwall R, Kumar G, Sharma B, Sarin S.
97. Kim J, Gong E, Seo M, et al. Timing of endoscopy in patients Post EVL (endoscopic variceal ligation) ulcer bleeding: A new
with upper gastrointestinal bleeding. Sci Rep. 2022;12(1):6833 classification and outcomes. Gastroenterol. 2017;152(5):S908
98. W Z, Xuang Y, Huiling XL, Z, Yuan L, Wang X, Dang T. Tim- 116. Jamwal KD, Post EVL. Bleeding Ulcers: A Pictorial Mini Re
ing of endoscopy for acute variceal bleeding in patients with view. Clin Cases Med. 2021;1(1):1006
cirrhosis (CHESS1905): A nationwide cohort study. Hepatol 117. Tripathi D, Therapondos G, Jackson E, Redhead D, Hayes P. The
Comm. 2023;7(5):e0152. role of the transjugular intrahepatic portosystemic stent shunt
99. ASGE Standards of Practice Committee, Ben-Menachem T, (TIPSS) in the management of bleeding gastric varices: clinical
Decker G, Early D, et al. Adverse events of upper GI endos- and haemodynamic correlations. Gut. 2005;51:270–274
copy. Gastrointest Endosc. 2012;76(4):707–718 118. Rinella M, Shah D, Vogelzang R, Blei A, Flamm S. Fundal
100. Koch D, Arguedas M, Fallon M. Risk of aspiration pneumo- variceal bleeding after correction of portal hypertension in
nia in suspected variceal hemorrhage: the value of prophylac- patients with cirrhosis. Gastrointest Endosc. 2003;58:122–127
tic endotracheal intubation prior to endoscopy. Dig Dis Sci. 119. Chirapongsathorn S, Manatsathit W, Farrell A, Suksamai A.
2007;52:225–2228 Safety and efficacy of endoscopic cyanoacrylate injection in the
101. Hayat U, Lee P, Ulah H, Servepalli S, Lopez R, Vargo J. management of gastric varices: a systematic review and meta-
Association of prophylactic endotracheal intubation in criti- analysis. JGH Open. 2021;5(9):1047–1055
cally ill patients with upper GI bleeding and cardiopulmonary 120. Ryan B, Stockbrugger R, Ryan J. A pathophysiologic, gastroen-
unplanned events. Gastrointest Liver Dis. 2016;86(3):500–509 terologic, and radiologic approach to the management of gastric
102. Chaudhuri D, Bishay K, Tandon P, et al. Prophylactic endotra- varices. Gastroenterology. 2004;126:1175–1189
cheal intubation in critically ill patients with upper gastrointes- 121. Sarin S, Jain A, Jain M, Gupta R. A randomized controlled trial
tinal bleed: a systematic review and meta-analysis. JGH Open. of cyanoacrylate versus alcohol injection in patients with isolated
2019;4(1):22–28 fundic varices. Am J Gastroenterol. 2002;97:1010–1015
103. Pasha S, Tarar Z, Chela H, McDermott A, Ghouri Y, Bechtold 122. Lo G, Lai K, Cheng J, Chen M, Chiang H. A prospective, ran-
M. Should prophylactic endotracheal intubation be performed domized trial of butyl cyanoacrylate injection versus band liga-
in upper GI bleeding? A meta-analysis. Am J Gastroenterol. tion in the management of bleeding gastric varices. Hepatology.
2022;117(10S):e446 2001;33:1060–1064
104. Park C, Park S, Jung J, et al. Clinical outcomes of sedation dur- 123. Tan P, Hou M, Lin H, et al. A randomized trial of endoscopic
ing emergency endoscopic band ligation for variceal bleeding: treatment of acute gastric variceal hemorrhage: N-butyl-
multicenter cohort study. Dig Endosc. 2020;32(6):894–903 2-cyanoacrylate injection versus band ligation. Hepatology.
105. Aziz M, Haghbin H, Gangwani M, et al. Erythromycin improves 2006;43:690–697
the quality of esophagogastroduodenoscopy in upper gas- 124. Lo G, Liang H, Chen W, et al. A prospective, randomized con-
trointestinal bleeding: a network meta-analysis. Dig Dis Sci. trolled trial of transjugular intrahepatic portosystemic shunt
2023;68(4):1435–1446 versus cyanoacrylate injection in the prevention of gastric
106. Manupeeraphant P, Wanichagool D, Songlin T, et al. Intravenous variceal rebleeding. Endosc. 2007;39:679–685
metoclopramide for increasing endoscopic mucosal visualization 125. Nadler J, Stankovic N, Uber A, et al. Outcomes in variceal
in patients with acute upper gastrointestinal bleeding: a multi- hemorrhage following the use of a balloon tamponade device.
center, randomized, double-blind, controlled trial. Sci Rep. 2024. Am J Emerg Med. 2017;35(10):1500–1502
https://​doi.​org/​10.​1038/​s41598-​024-​57913-2 126. Monescillo A, Martinez-Lagares F, Ruiz-del-Arbol L, et al.
107. Patra BR, Harindranath S, Ansari AA, et al. Utility of gastros- Influence of portal hypertension and its early decompression by
copy in the left lateral semi-recumbent position: A blood-free TIPS placement on the outcome of variceal bleeding. Hepatol.
coup! Endosc Int Open. 2023;11(12):1168–1174 2004;40:793–801
108. Sarin S, Govil A, Jain A, et al. Prospective randomized trial 127. Nicoara-Farcau O, Han G, Rudler M, Angrisani D, Monescillo
of endoscopic sclerotherapy versus variceal band ligation for A, Torres F, et al. Pre-emptive TIPS in high-risk acute variceal
esophageal varices: influence on gastropathy, gastric varices and bleeding. An updated and revised individual patient data meta-
variceal recurrence. J Hepatol. 1997;26(4):826–832 analysis. Hepatology. 2024;79(3):624–635
109. Ferrari A, De-Paulo G, De-Macedo C, Araújo I, Della Libera E. 128. Lee EW, Saab S, Eghbalieh N, Ding PX, Jeon UB, Ohm JY,
Efficacy of absolute alcohol injection compared with band liga- et al. Coil or plug-assisted retrograde transvenous obliteration
tion in the eradication of esophageal varices. Arq Gastroenterol. (CARTO/PARTO) for treating portal hypertensive variceal
2005;42(2):72–76 bleeding: A multicenter, real-world 10-year retrospective
110. Kalista K, Hanif S, Nababan S, Lesmana C, Hasan I, Gani R. The study. Hepatology. 2025 Feb 5. https://​doi.​org/​10.​1097/​HEP.​
clinical role of endoscopic ultrasound for management of bleed- 00000​00000​001255. Online ahead of print.
ing esophageal varices in liver cirrhosis. Case Rep Gastroenterol. 129. Bernard B, Cardanel J. Prognostic significance of bacterial
2022;16(2):295–300 infection in bleeding cirrhotic patients: a prospective study.
111. Rivory J, Pioche M, Dumortier J, et al. Transesophageal endo- Gastroenterol. 1995;108(1828–1834).
scopic ultrasound-guided coil and cyanoacrylate treatment of 130. Goulis J, Armonis A, Patch D, et al. Bacterial infection is inde-
challenging esophageal varices bleeding associated with CREST pendently associated with failure to control bleeding in cir-
syndrome ulcerative esophagitis. Endosc. 2022;54(12):e761-762 rhotic patients with gastrointestinal hemorrhage. Hepatology.
112. Oleas R, Robles-Medranda C. Endoscopic treatment of gastric 1998;27:1207–1212
and ectopic varices. Clin Liver Dis. 2022;26(1):39–50 131. Bernard B, Grange J, Khac EA, X, Opolon P, Poynard T. Anti-
113. Huang Y, Zhang W, Xiang H, et al. Treatment strategies in biotic prophylaxis for the prevention of bacterial infections
emergency endoscopy for acute esophageal variceal bleeding
 Hepatology International

in cirrhotic patients with gastrointestinal bleeding: A meta- 141. Hidajat N, Stobbe H, Hosten N, et al. Transjugular intrahepatic
analysis. Hepatol. 1999;29:1655–1661. portosystemic shunt and transjugular embolization of bleed-
132. Avgerinos A, Armonis A, Stefanidis G, et al. Sustained rise ing rectal varices in portal hypertension. Am J Roentgenol.
of portal pressure after sclerotherapy, but not band liga- 2002;179:362–363
tion, in acute variceal bleeding in cirrhosis. Hepatology. 142. Garcia-Tsao G, Sanyal A, Grace N, Carey W. Prevention and
2004;39:1623–1630 management of gastroesophageal varices and variceal hemor-
133. Villanueva C, Piqueras M, Aracil C, et al. A randomized con- rhage in cirrhosis. Hepatology. 2007;46(3):922–938
trolled trial comparing ligation and sclerotherapy as emergency 143. Goulis J, Patch D, Burroughs A. Bacterial infection in the patho-
endoscopic treatment added to somatostatin in acute variceal genesis of variceal hemorrhage. Lancet. 1999;353:139–142
bleeding. J Hepatol. 2006;45:560–567 144. Villanueva C, Ortiz J, Sabat M, et al. Somatostatin alone or com-
134. Moitinho E, Escorsell A, Bandi J, et al. Prognostic value of bined with emergency sclerotherapy in the treatment of acute
early measurements of portal pressure in acute variceal bleed- esophageal variceal bleeding: a prospective randomized trial.
ing. Gastroenterol. 1999;117:626–631 Hepatology. 1999;30:384–389
135. Ready J, Robertson A, Goff J, Rector W. Assessment of the risk 145. Levacher S, Letoumelin P, Pateron D, Blaise M, Lapandry C,
of bleeding from esophageal varices by continuous monitoring Pourriat J. Early administration of terlipressin plus glyceryl trini-
of portal pressure. Gastroenterol. 1991;100:1403–1410 trate to control active upper gastrointestinal bleeding in cirrhotic
136. Abraldes J, Villanueva C, Banares R, et al. Hepatic venous patients. Lancet. 1995;346:865–868
pressure gradient and prognosis in patients with acute variceal 146. D’Amico G, Pagliaro L, Bosch J. Pharmacological treatment of
bleeding treated with pharmacologic and endoscopic therapy. portal hypertension: an evidence-based approach. Semin Liver
J Hepatol. 2008;48:229–236 Dis. 1999;19:475–505
137. Villanueva C, Planella M, Aracil C, Lopez-Balaguer JG, B, 147. Schneider BL, Bosch J, de Franchis R, Emre SH, Groszmann RJ,
Minana J, et al. Hemodynamic effects of terlipressin and high Ling SC, et al. Portal hypertension in children: expert pediatric
somatostatin dose during acute variceal bleeding in nonre- opinion on the report of the Baveno V Consensus Workshop on
sponders to the usual somatostatin dose. Am J Gatroenterolol. methodology of diagnosis and therapy in portal hypertension. J
2005;100:623–630. Pediatr Gastroenterol Nutr. 2012;55(4):491–495
138. Helmy A, Al Kahtani K, Al FM. Updates in the pathogenesis, 148. Bello FPS, Cardoso S, Tannuri AC, Preto-Xamperlini M, Sch-
diagnosis and management of ectopic varices. Hepatol Int. vartsman C, Farhat SCL. Acute upper gastrointestinal bleeding
2008;2:322–334 due to portal hypertension in children: What is the best timing
139. Kinkhabwala M, Mousavi A, Iyer S, Adamsons R. Bleeding of endoscopy? Dig Liver Dis. 2022;54(1):63–68
ileal varicosity demonstrated by transhepatic portography. Am
J Roentgenol. 1977;129:514–516 Publisher's Note Springer Nature remains neutral with regard to
140. Macedo T, Andrews J, Kamath P. Ectopic varices in the gastro- jurisdictional claims in published maps and institutional affiliations.
intestinal tract: short- and long-term outcomes of percutaneous
therapy. Cardiovasc Intervent Radiol. 2005;28:178–184

Authors and Affiliations

Cosmas Rinaldi Adithya Lesmana1,2,3 · Akash Shukla4,5 · Ashish Kumar6 · Shalimar7 · Xiaolong Qi8 ·
Rino Alvani Gani1 · Ze‑Hao Zhuang9 · Abdul Kadir Dokmeci10 · Gin Ho Lo11 · Hitoshi Maruyama12 · Ji‑Dong Jia13 ·
Anand V. Kulkarni14 · Jason Chang15 · Necati Ormeci16 · Gamal Shiha17 · Hasnain Ali Shah18 · Jose D. Sollano19 ·
Sahaj Rathi20 · Tan Soek Siam21 · George K. Lau22 · Rungsun Rerknimitr23 · Ming‑Chih Hou24 · Juferdy Kurniawan1 ·
Guohong Han25 · Amar Mukund26 · Sanjay Saran Baijal27 · Shiv. Kumar Sarin26

7
* Cosmas Rinaldi Adithya Lesmana All India Institute of Medical Sciences, New Delhi, India
[email protected] 8
Liver Disease Center of Integrated Traditional Chinese
1 and Western Medicine, Department of Radiology,
Hepatobiliary Division, Department of Internal Medicine,
Zhongda Hospital, Medical School, Southeast University,
Dr. Cipto Mangunkusumo National General Hospital,
Nurturing Center of Jiangsu Province for State Laboratory
Medical Faculty Universitas Indonesia, Jakarta, Indonesia
of AI Imaging and Interventional Radiology (Southeast
2
Digestive Disease and Gastrointestinal Oncology Center, University), Nanjing, China
Medistra Hospital, Jakarta, Indonesia 9
The Second Hospital of Fujian Medical University,
3
Gastrointestinal Cancer Center, MRCCC Siloam Semanggi Quanzhou, People’s Republic of China
Hospital, Jakarta, Indonesia 10
Department of Gastroenterology, Faculty of Medicine,
4
Department of Gastroenterology, Seth GS Medical College Ankara University, Ankara, Turkey
and KEM Hospital, Mumbai, India 11
E-Da Hospital, I-Shou University, Kaohsiung, Taiwan
5
Department of Hepatology, Sir HN Reliance Foundation 12
Department of Gastroenterology, Juntendo University,
Hospital, Girgaon, Mumbai, India
Tokyo, Japan
6
Institute of Liver, Gastroenterology and Pancreatico‑Biliary 13
Liver Research Center, Beijing Friendship Hospital, Capital
Sciences, Sir Ganga Ram Hospital, New Delhi, India
Medical University, Beijing, China
Hepatology International

14 22
Department of Hepatology, Asian Institute Humanity and Health Medical Group, Hong Kong, HKSAR,
of Gastroenterology, Hyderabad, India China
15 23
Department Gastroenterology and Hepatology, Singapore Division of Gastroenterology, Department of Medicine,
General Hospital, Singapore, Singapore Faculty of Medicine, Excellence Center for Gastrointestinal
16 Endoscopy, King Chulalongkorn Memorial Hospital,
The Department of Gastroenterohepatology, Istanbul Health
Bangkok, Thailand
and Technology University, Istanbul, Turkey
24
17 National Yang-Ming Chiao-Tung University School
Department of Internal Medicine, Faculty of Medicine,
of Medicine, Taipei, Taiwan R.O.C.
Mansoura University, Mansoura, Egypt
25
18 Department of Liver Diseases and Digestive Interventional
Aga Khan University, Karachi, Pakistan
Radiology, Digestive Diseases Hospital, Xi’an International
19
University of Santo Tomas Hospital, Manila, Philippines Medical Center Hospital, Xi’an, China
20 26
Post Graduate Institute of Medical Education and Research, Institute of Liver and Biliary Sciences, New Delhi, India
Chandigarh, India 27
Medanta Hospital, Gurugram, India
21
Department of Medicine, Hospital Selayang, Bata Caves,
Selangor, Malaysia