E n d o s c o p i c Ma n a g e m e n t o f

P o r ta l Hy p er t en s i o n – rel a te d
Bleeding
Andrew Nett, MD*, Kenneth F. Binmoeller, MD

KEYWORDS
 Varices  Portal hypertension  GI bleed  Therapeutic endoscopy

KEY POINTS
 Portal hypertension–related bleeding can be catastrophic and fatal. Multidisciplinary man-
agement is necessary, potentially involving medical therapies, endoscopic intervention,
and/or percutaneous or surgical portosystemic shunt creation.
 A wide variety of endoscopic and endosonographic therapies are available for hemostatic
and prophylactic intervention. The appropriate therapy depends on the vascular lesion
and its location.
 Endoscopic ultrasound (EUS) enhances the endoscopic management of gastric and
ectopic varices. EUS-guided intravascular coil deployment may mitigate the risk of
embolic complications associated with tissue adhesive injection.

INTRODUCTION

Gastrointestinal (GI) bleeding as a sequela of portal hypertension can be catastrophic

and fatal. The overall approach to portal hypertension and related bleeding is multidis-
ciplinary, potentially involving medical interventions and/or percutaneous or surgical
portosystemic shunting for portal hypertension reduction. In addition, while doing
nothing to correct underlying portal hypertension, endoscopic and endosonographic
therapy plays a critical role in management—both for the achievement of hemostasis
during active bleeding and for prevention of initial and recurrent bleeding.

PORTAL HYPERTENSION

Portal hypertension results from several disease states affecting the prehepatic, intra-
hepatic, or posthepatic circulation. In the Western world, sinusoidal hypertension from

Disclosure Statement: The authors have nothing to disclose.

Interventional Endoscopy Services, California Pacific Medical Center, 2351 Clay Street, 6th Floor
Suite 600, San Francisco, CA 94115, USA
* Corresponding author.
E-mail address: Andrewnett83@Gmail.com

Gastrointest Endoscopy Clin N Am 29 (2019) 321–337

https://doi.org/10.1016/j.giec.2018.12.006 giendo.theclinics.com
1052-5157/19/Published by Elsevier Inc.
322 Nett & Binmoeller

cirrhosis is the most common cause of portal hypertension.1 In cirrhosis, abnormal si-
nusoidal blood flow occurs due to structural aberrations, including scarring, regener-
ative nodules, and microvascular clotting, cause increased intrahepatic resistance
and resultant portal hypertension.2 Intrahepatic resistance is also elevated by
increased vascular tone related to higher levels of local vasoconstrictors and reduced
nitric oxide levels from endothelial dysfunction. Myofibroblasts form from hepatic
stellate cells in response to inflammation and cytokine release induced by hepatocyte
injury. These myofibroblasts also contract within the space of Disse, adding to
increased intrahepatic resistance.1
When portal pressure in sufficiently elevated to be clinically significant (hepatic vein
portal gradient [HPVG] >10 mm Hg), portosystemic collaterals develop. These collat-
erals arise with recannulation of embryonic vascular channels, reversal of flow within
adult veins, or from neoangiogenesis.3 Collateralization does not effectively relieve
portal hypertension in cirrhotic patients, however. Increased splanchnic nitric oxide
production results in splanchnic vasodilation and increased splanchnic blood flow,
contributing to a persistent portal hypertensive state despite portosystemic collateral
development. Back pressure caused by this portal hypertension thus transmits
through these collaterals to perforating veins and the submucosal and subepithelial
vessels they supply, which is where clinically significant varices may form. Esopha-
geal and gastric varices are most common, but ectopic varices can develop any-
where in the GI tract.

ESOPHAGEAL VARICES

Esophageal varices are present in 30% to 40% of patients with compensated

cirrhosis, but in up to 85% of those with decompensated cirrhosis.2 The rate of vari-
ceal hemorrhage is approximately 10% to 15% per year with an associated 6-week
mortality of 15% to 25%. Recurrent variceal hemorrhage occurs in 60% of patients
without treatment.
The thoracic esophagus typically drains into the azygous and hemiazygos systemic
venous systems.3 In contrast, the abdominal esophagus typically drains to the portal
system via the left gastric vein with lesser drainage to the systemic vascular system via
superior and inferior phrenic veins into the inferior vena cava. With reversal of flow, it is
the left gastric vein that usually acts as the afferent vessel supplying esophageal vari-
ces in cirrhotic patients.

Scenarios for Endoscopic Therapy

Primary prophylaxis
Endoscopic management may be appropriate for primary prophylaxis against
bleeding of esophageal varices. The role for endoscopic intervention depends on
the severity of underlying portal hypertension, variceal size, and whether or not
decompensated cirrhosis is present. Patients with mild portal hypertension
(HPVG >5 mm Hg but <10 mm Hg) do not develop varices.2 Once clinically significant
portal hypertension (HPVG >10 mm Hg) is present, however, variceal development
may follow and there is potential for intervention prior to complication by variceal
hemorrhage.
For this scenario—described as primary prophylaxis—intervention may be either
medical or endoscopic. Endoscopic therapy is specifically considered in patients
with evidence of increased risk of variceal bleed. This population includes all patients
with decompensated cirrhosis and those with compensated cirrhosis with either me-
dium to large varices (>5 mm). In patients with medium to large varices, endoscopic
 Endoscopic Management of Portal Hypertension 323

therapy with variceal ligation has been associated with decreased rates of variceal
hemorrhage compared with medical therapy with a nonselective b-blocker.
Per the 2017 American Association for the Study of Liver Diseases guidance state-
ment, small (<5 mm) but high-risk varices (small varices in decompensated patients or
small varices with red wale signs) warrant nonselective b-blockers as preferential ther-
apy over endoscopic intervention.

Secondary prophylaxis
For treatment of active variceal bleeding or for secondary prophylaxis after a variceal
bleed, endoscopic intervention is indicated.

Technique/Method of Endoscopic Therapy

Endoscopic variceal ligation
The first-line endoscopic therapy for management of esophageal varices is endo-
scopic variceal ligation (EVL). EVL was introduced as a therapeutic option in 1986.4
In EVL, a cap-assisted ligation device mounted on the scope is used to deploy an
elastic ring ligature around a target varix suctioned into the cap. After suctioning of
the varix into the cap, clockwise turning of a firing device attached at the external bi-
opsy valve port results in ligature firing.
Particularly in the setting of scar tissue from prior therapy, adequate suctioning of
the varix into the cap to ensure varix entrapment may be difficult. Prolonged, steady
suctioning while gently torqueing the scope back and forth often induces more tissue
into the cap. Release of suctioning prior to ligation deployment should generally be
avoided, because this action may be enough to precipitate active bleeding. Once
deployed successfully, the ligature entraps the mucosal and submucosal esophagus
layers, causing varix strangulation for immediate hemostasis followed by intravascular
thrombi formation, necrosis, and consequent fibrosis for variceal obliteration.
Active bleeding may be present that significantly impairs visualization. Water infu-
sion can help obtain visualization of the bleeding source. If necessary, blind banding
at the gastroesophageal (GE) junction may reduce bleeding enough for visualization
and appropriate targeting of subsequent bands.
Ligature deployment for decompression along a variceal column should be per-
formed distal to proximal because a caudal to rostral pressure gradient is typically pre-
sent within the varix. In addition, the band-varix complex limits readvancement of the
endoscope distally due to partial occlusion of the lumen and risk of band displace-
ment. In the setting of acute variceal hemorrhage, banding should be preferentially tar-
geted toward the culprit variceal column evidenced by stigmata of hemorrhage. In
prophylactic EVL, red wale signs may also direct band placement. Generally, helical
placement of bands moving distal to proximal allows application of the maximum
number of bands while avoiding overlapping circumferential placement.5
Bands are typically placed along the distal 8 cm segment of the esophagus. This
range is enough to target the palisade drainage and perforating zones. The palisade
drainage zone (extending from the GE junction to 2–3 cm superiorly) is an azygous-
portal watershed where veins in the proximal gastric submucosa penetrate through
the muscularis mucosae at the GE junction and then course superficially within the
lamina propria. The perforating zone extends from 3 cm to 5 cm from the GE junction,
within which perforating veins extend from extrinsic paraesophageal venous plexuses,
penetrate the muscularis propria, and supply intrinsic submucosal veins.5–7 Due to
their organization and response to portal hypertension, these areas are suggested crit-
ical areas of variceal development and rupture.7 More proximal placement may have
less efficacy and may serve only to exacerbate postligation retrosternal discomfort.
324 Nett & Binmoeller

Multiple bands are typically placed until varices appear decompressed or until neces-
sary placement becomes too proximal.
Efficacy of endoscopic variceal ligation
EVL is effective in immediate control of active variceal hemorrhage in approximately
90% of cases. Complications after EVL occur approximately 2% to 20% of the time
and include transient dysphagia, retrosternal pain, esophageal stricture, ulcerations,
perforation, and infection.8 Rebleeding, sometimes massive, can also occur, either
from recurrent variceal rupture or from postligation ulceration.
Ulcer management
After EVL, the ligature bands may stay in place for a range of 3 days to 7 days. An ulcer
remains that heals within 2 weeks to 3 weeks.9 If thrombus formation is incomplete
when the ligature band sloughs off, postligation ulcer bleeding may occur. Overall
the risk of post-EVL ulcer bleeding is 3.6% to 15%.10 Such bleeding can often be
managed conservatively with PPI therapy and supportive care. Nonendoscopic man-
agement with emergency transjugular intrahepatic portosystemic shunt (TIPS) or
esophageal balloon tamponade may be pursued in cases of massive hemorrhage. Op-
tions for endoscopic treatment of postligation bleeding include repeat band ligation,
endoscopic variceal obturation with cyanoacrylate injection, self-expandable metal
stent placement, or hemostatic powder spray application.10–12
Interventions Other Than Endoscopic Variceal Ligation
Injection sclerotherapy
Prior to EVL, injection sclerotherapy was the first endoscopic treatment proved supe-
rior to balloon tamponade or vasoconstrictor administration in the management of
esophageal varices. Unfortunately, complications occur in up to 40% of patients
receiving sclerotherapy (esophageal ulceration, stricture, perforation, pulmonary
thrombus, pleural effusion, hemothorax, mediastinitis, pericarditis, pneumothorax,
renal dysfunction, and death).8,13 Mortality related to this therapy occurs in 1% to
2%.14
Several studies have shown that band ligation is superior to endoscopic injection
sclerotherapy. A 1995 meta-analysis of 7 randomized trials published in 2015 demon-
strated EVL resulted in lower rebleeding, mortality, and complications as well as the
need for fewer treatment sessions.15 Similarly, Dai and colleagues14 reported a
meta-analysis of 14 studies involving 1236 patients treated for active esophageal var-
iceal hemorrhage. EVL resulted in significantly lower rebleeding rates (27.7% vs
33.1%), higher eradication rates, lower complications, and no difference in mortality.
A meta-analysis reviewing variceal band ligation alone versus band ligation plus
sclerotherapy as therapy for secondary prophylaxis found that the addition of therapy
did not improve rebleeding, the number of endoscopic sessions required for variceal
obliteration, procedure-related complications, or mortality. Adding sclerotherapy to
band ligation was associated, however, with higher rates of esophageal structuring.16
Despite its inferiority, injection sclerotherapy is still a potential option applied when
EVL is technically difficult.17 As discussed previously, significant active bleeding may
impair visualization, complicating band ligation. Scar tissue may also prevent
adequate suctioning of varices into a cap to achieve band ligation. Sclerosant injection
is an option in such cases.
Agents with reported use consist of sodium tetradecyl sulfate (Food and Drug
Administration approved), ethanolamine oleate, sodium morrhuate, polidocanol, or
absolute alcohol. Intravariceal injection may be performed just distal to the site of
bleeding or paravariceal injection may be performed immediately adjacent to a varix.
 Endoscopic Management of Portal Hypertension 325

The sclerosant precipitates inflammation and thrombosis. After injection near the
bleeding site, injections are performed starting at all varices at the GE junction, with
proximal injections at 2-cm intervals, extending up to 5 cm to 6 cm from the GE
junction.8

Metal stents
In cases of refractory or uncontrolled variceal bleeding, which occurs approximately
10% of the time despite EVL, balloon tamponade and TIPS are emergent salvage ther-
apies. Balloon tamponade may be performed with placement of a Sengstaken-
Blakemore tube. Although this method is helpful for hemostasis, it is only temporizing
with a high risk of bleeding recurrence after deflation, which occurs in up to 50% of
patients. There is also significant risk of other major complications including aspiration
pneumonia and esophageal perforation (rate of up to 30%).18 Covered self-expanding
metal stent placement provides an endoscopic alternative to salvage therapy.
A specific stent system has been designed for treatment of acute variceal hemor-
rhage in which a 25-mm  13.5-cm fully covered, self-expanding nitinol metal stent
is used (SX-ELLA Danis Stent, Ella-CS, Hradec Králové, Czech Republic). This stent
may be deployed without fluoroscopy or endoscopy although endoscopic guidance
has been used in most reported cases.18–20 The stent has atraumatic edges and is
left in place for up to 14 days prior to removal. The stent thus can serve as bridge ther-
apy for subsequent repeat EVL or TIPS.
A randomized controlled trial (RCT) comparing use of this stent versus balloon tam-
ponade showed stenting resulted in significantly increased therapeutic success (66%
vs 20%), decreased rebleeding at 15 days (77% vs 52%) and decreased adverse
events (8% vs 52%) but no difference in 6-week survival.21 In a 2017 meta-analysis
examining 80 total cases in which esophageal stenting was performed for refractory
variceal bleeding, a therapeutic success rate of 93.9% was reported with a 13.2%
rebleed rate, no stent-related complications, and 21.6% stent migration rate. Mortality
was 34.5% (only 12.6% died from uncontrolled bleeding).20

Argon plasma coagulation

A prospective, randomized trial compared EVL followed by argon plasma coagulation
(APC) and EVL alone for esophageal variceal management. Significantly lower recur-
rence of esophageal varices was noted at 24 months after treatment when APC was
also used (49.6% vs 74.2%). More recent RCTs examined a role for APC in secondary
prophylaxis. Cipolletta and colleagues22 in 2002 randomized 30 cirrhotics with a his-
tory of acute variceal hemorrhage and subsequent EVL eradication of varices to either
observation versus APC. APC was applied to the entire circumferential mucosa span-
ning 4-cm to 5-cm proximal to the GE junction in 1 session to 3 sessions at weekly in-
tervals. No significant complications were reported although APC did cause transient
fever in 13 of 16 patients and 8 of 13 had dysphagia or retrosternal discomfort; 0%
recurrence occurred in patients after APC therapy when followed with endoscopy
for every 3 months for up to 28 months (9–28 mo). Meanwhile, recurrence occurred
in 43% of the control group, with a 7% recurrent bleeding rate.
A 2017 meta-analysis by Li and colleagues23 compared the safety and efficacy of
EVL alone versus EVL 1 APC for secondary prophylaxis of esophageal variceal
bleeding. Across 4 RCTs, combination therapy did show significantly lower variceal
recurrence rates (relative risk 0.09). No difference in bleeding recurrence or mortality
was present. Fever occurred more often after combination therapy. In addition, in
2012, 60 patients with large varices were randomized to primary prophylactic therapy
with either EVL alone or EVL plus postvariceal obliteration APC for 1 session. In this
326 Nett & Binmoeller

study, APC did not result in any improvement in variceal recurrence at 7-month follow-
up. Neither group had variceal bleeding.

Clipping and endoloops

Prospective studies have suggested either endoscopic clipping or endoloop ligation
may achieve comparable rates of initial effective hemostasis and rebleeding
compared with EVL.24,25 Clipping resulted in a decreased number of treatment ses-
sions until obliteration whereas endoloop ligation and EVL were comparable in this
regard. Data for both techniques are limited, however, and both studies were non-
randomized. With clipping, the risk of precipitating significant bleeding from variceal
puncture by clip prongs exists and the cost of multiple clips seems exorbitant
compared with that of the band ligation device.

Tissue adhesive injection

The injection of cyanoacrylate has been compared with EVL for the initial treatment of
medium to large esophageal varices in an RCT published by Santos and colleagues.26
The 2 therapies resulted in similar rates of variceal eradication, major complications,
and mortality when followed over 6 months, but cyanoacrylate resulted in more minor
complications and variceal recurrence (33% vs 57% recurrence).
Adhesive injection may also be performed for active variceal bleeding. Tissue ad-
hesives used in acute variceal hemorrhage may achieve hemostasis in a high number
of patients (94.2% in a study of 133 cirrhotic patients with 52 with active and 81 with
recent bleeding).27 In class C cirrhotic patients, Ribeiro and colleagues28 reported in
2015 a retrospective study involving 63 patients admitted with acute esophageal var-
iceal hemorrhage that n-butyl 2-cyanoacryle injection achieved hemostasis in
approximately 75% of patients. Only 46% of patients were able to go 6 weeks after
treatment without bleeding, however. In 2011, a small RCT compared EVL and injec-
tion of n-butyl cyanoacrylate in acute variceal bleeding.29 Regardless of the initial
therapy, follow-up EVL was performed within 2 weeks until varices were completely
eradicated. Both interventions were comparable in achieving hemostasis. Although
there was no difference in initial hemostasis, there was a suggestion of higher
rebleeding in patients receiving n-butyl cyanoacrylate injection (although not statisti-
cally significant—13.6% vs 4.7%; n 5 0.60692).
Overall, there is no definitive evidence supporting use of cyanoacrylate injection for
management of bleeding varices or for variceal bleed prophylaxis. Tissue adhesive in-
jection may be considered in lieu of or in conjunction with sclerotherapy, however,
when EVL is technically challenging. Sclerotherapy is, of course, a second resort to
EVL for treatment of bleeding esophageal varices, but an RCT did show n-butyl cyano-
acrylate and sclerosant injection resulted in lower rebleeding and mortality compared
with sclerotherapy alone.30 Other studies have also shown combination sclerosant
and n-butyl cyanoacrylate injection results in lower rebleeding, mortality, and minor
complications compared with sclerotherapy alone.31,32

GASTRIC VARICES

Gastric varices are the second most common cause of GI bleeding in patients with
portal hypertension. Gastric variceal bleeding, although less prevalent, has the pro-
pensity to be more severe, associated with higher transfusion requirement and with
higher morbidity and mortality compared with esophageal variceal bleeding.33,34
Afferent vessels supplying gastric varices typically consist of the left gastric, posterior
gastric, and short gastric veins. Varices arising with left gastric supply develop in the
cardia and those arising from the short and posterior gastric veins form in the fundus.
 Endoscopic Management of Portal Hypertension 327

Gastric varices anastomose with the systemic circulation most commonly through
esophageal varices into the superior vena cava. When gastric varices are isolated,
they typically have afferent venous drainage through posterior or short gastric veins
while efferent drainage is through a gastric—splenorenal shunt and the inferior vena
cava via the inferior phrenic veins.
The Sarin classification system categorizes gastric varices based on whether or not
they are contiguous with esophageal varices and their location in the stomach.
Gastroesophageal varices type 1 (GOV1) and GOV2 varices are contiguous with
esophageal varices extending either into the lesser curvature (GOV1) or the fundus
along the greater curvature (GOV2).3,6 These varices, also known as junctional varices,
share the pathophysiology of esophageal varices, arising from the left gastric vein and
originating in the lamina propria.35
Isolated Gastric Varices type 1 (IGV1) and IGV2 varices are fundal-type varices distinct
from esophageal varices formed in the cardia (IGV1) or outside of the cardia and fundus,
usually around the antrum or pylorus (IGV2).3,6 They arise from the short and posterior
gastric veins, originating in the submucosa. Bleeding risk is significantly higher for fundal
varices (IGV1 77% and GOV2 55%) than either GOV1 or ectopic varices (10%).36 Extra-
hepatic portal vein obstruction more commonly results in IGV1 varices whereas cirrhotic-
related portal hypertension more commonly results in GOV2 varices.34
The Sarin classification system aligns with therapeutic distinctions. Specifically,
GOV1 varices may be managed the same as esophageal varices. In this scenario,
obliteration of esophageal varices with EVL frequently results in disappearance of
the gastric varices as well in approximately 60% of patients.36

Sclerotherapy
Endoscopic sclerosant injection is not a good option in management of gastric varices
due to association with high rates of complication including gastric ulceration, perfo-
ration, and rebleeding (37%–53%).36,37 Furthermore, in patients with IGV1 varices, it is
associated with low rates of variceal obliteration (41%) and high rates of rebleeding
may occur (37% to 89%).38,39 In 2002, Sarin and colleagues40 performed an RCT
comparing cyanoacrylate injection versus alcohol-based sclerotherapy in 37 patients
with isolated fundal gastric varices (17 of 37 presented with active bleeding). Cyano-
acrylate was more efficacious than alcohol in immediate hemostasis (89% vs 62%)
and variceal obliteration (100% vs 44%), achieving obliteration significantly quicker
(2 weeks vs 4.7 weeks).

Endoscopic Variceal Ligation Versus Tissue Adhesive Injection

As discussed previously, GOV1 varices specifically may be managed the same as
esophageal varices. For other varix types, however, EVL can result in severe bleeding
and is not advised for patients with IGV varices, which arise in the submucosa. The
presence of thicker overlying mucosa within the stomach and the larger size of IGVs
both make it difficult to fully entrap a gastric varix with suctioning into the banding liga-
tion cap. If a band ligature does not fully capture the deeper gastric variceal wall,
thrombotic obliteration of the lumen may not occur and postligation ulceration with
sloughing of the ligature can result in severe hemorrhage.2,41 Instead of band ligation,
the current worldwide standard for treatment of IGVs is glue injection. Several studies
support this practice.
Soehendra and colleagues42 first reported the use of n-2-butyl-cyanoacrylate for
endoscopic injection of gastric varices in 1986. This method is capable of achieving
58% to 100% hemostasis, with a rebleed rate ranging from 0% to 40%.43 In 2013,
Tantau and colleagues44 prospectively compared n-butyl 2-cyanoacrylate injection
328 Nett & Binmoeller

with EVL in acute gastric variceal bleeding. Both treatments obtained initial hemosta-
sis in a high percentage of patients. Rebleeding, however, was significantly more
frequent after EVL (72% vs 32%; P 5 .03) and the rebleeding-free period was larger
after glue injection (P 5 .006).
Subsequent RCTs have buttressed the superiority of cyanoacrylate injection. A
meta-analysis reviewing treatment of acute gastric variceal bleeding compared
cyanoacrylate injection versus other endoscopic intervention. Analysis of 3 RCTs
showed that EVL and cyanoacrylate injection achieve comparable rates of initial
hemostasis, but EVL therapy carries a significantly higher risk of rebleeding.41 Treat-
ments were similar in bleeding control, complications, and bleeding-related mortality
although cyanoacrylate injection was found superior to EVL in prevention of rebleed-
ing. The largest RCT included in this meta-analysis examined only GOV1 varices,
expected to overestimate the efficacy of EVL applied to cardiofundal varices
(IGV1/GOV2).45

Endoscopic ultrasound–Guided Tissue Adhesive Injection

Endoscopic ultrasound (EUS) can enhance endoscopic management of gastric vari-
ces in several regards.46 EUS has been shown to increase the detection rate of gastric
varices 6-fold in cirrhotic patients (79% vs 12.5%).47 Due to their deeper intramural
location, gastric varices may be overlooked as gastric folds on direct endoscopy. After
therapy, Doppler assessment is valuable in confirming obliteration of vascular flow
through a variceal complex, which is important because residual flow has prognostic
implications regarding rebleeding risk.48 In cases of significant hemorrhage, torrential
active bleeding or large clots can prevent adequate visualization for identification of a
culprit gastric varix and subsequent cyanoacrylate injection. EUS enables guidance of
therapy without direct endoscopic visualization.46 EUS may also direct therapy more
precisely into a perforating vessel with preceding contrast injection showing whether
the feeding vessel is efferent or afferent, potentially reducing embolization risk by
enabling use of a smaller volume of cyanoacrylate.49 If endoscopic therapy fails,
EUS provides additional information regarding portal vein and splenic vein patency,
helping to assess candidacy for TIPS versus balloon-occluded retrograde transve-
nous obliteration.

Complications of tissue adhesive injection

Technical complications that may arise during cyanoacrylate injection include
adhesive-related endoscope damage or needle entrapment in a varix with adhesive
curation/hardening. Clinical complications from cyanoacrylate injection typically
involve the risk of adhesive systemic embolization with primary concern for pulmonary
embolism or paradoxic embolic events, such as cerebral embolism in patients with a
patent foramen ovale. Coronary and splenic artery embolism also may occur as well
as splenic and renal vein thrombosis. Sepsis has also resulted after glue embolization
with the embolus serving as a nidus of infection. Visceral fistulization from the stomach
into the pleura or mediastinum also may occur after unintentional paravariceal
injection.50
Embolization with cyanoacrylate injection actually may be alarmingly common
although fortunately typically inconsequential. Routine CT imaging after n-butyl
2-cyanoacrylate diluted with 1:1 lipidiol injection shows that asymptomatic glue
embolization may occur in approximately 50%.51 Clinically significant embolization,
in contrast, is fairly rare but when embolization is substantial, complications may be
catastrophic with resultant death.52 Embolization risk may be enhanced with exces-
sively rapid cyanoacrylate injection, large-volume glue injection, rapid variceal flow
 Endoscopic Management of Portal Hypertension 329

rate, or overdilution of n-butyl 2-cyanoacrylate with lipidiol, which prolongs

hardening.41

Tissue adhesive selection

Although most studies involving cyanoacrylate injection of gastric varies use n-butyl
2-cyanoacrylate, several cyanoacrylate monomers are available. n-Butyl 2-cyanoac-
rylate has a rapid polymerization time. Therefore, it is typically diluted with lipidiol to
avoid premature intraneedle hardening or needle entrapment within the varix.41 Water
and not saline should be flushed through the needle, because saline accelerates poly-
merization. 2-Octyl cyanoacrylate, in contrast, has a slower polymerization time
limiting risk of premature hardening and obviating need for lipidiol dilution, which com-
plicates injection due to its viscosity. The needle can also be flushed with saline when
2-octyl-cyanoacrylate is used. Of course, 2-octyl-cyanoacrylate may augment the risk
of distant embolization if rapid injection is performed.

Glue-coil Embolization
In further efforts to mitigate potentially devastating embolic complications after cyano-
acrylate injection, Binmoeller and colleagues39 first performed ex vivo analysis of the
utility of concomitant local coil and glue injection in the treatment of gastric varices.
Initial deployment of intravascular coils provides a scaffold for glue polymerization
and fixation to prevent inadvertent glue embolization. To test this hypothesis, in
ex vivo study, a coil was deployed into a container of heparinized blood; 1 mL of
cyanoacrylate was then injected with immediate adherence of the glue to the coil’s
synthetic fibers. The coil and its adherent glue were then removed from the container
and no free glue remained (Fig. 1).
Pioneering this combined injection technique in humans, Binmoeller and col-
leagues53 first reported EUS-guided glue-coil embolization in a case involving massive
hemorrhage refractory to standard cyanoacrylate injection. Subsequently, Binmoeller
and colleagues reported a series of 30 patients with large gastric varices with active or

Fig. 1. Glue fixated to coil scaffolding after injection into container of heparinized blood.
(From Weilert F, Binmoeller KF. Endoscopic management of gastric variceal bleeding. Gastro-
enterol Clin North Am 2014;43(4):812; with permission.)
330 Nett & Binmoeller

recent (<1 week) bleeding and poor candidacy for TIPS. After endosonographic visu-
alization of the gastric varices, transesophageal, transcrural deployment of a coil into a
gastric variceal lumen was performed using the 19-gauge echoendoscope followed
by injection of 1 mL of 2-octyl cyanoacrylate. Observation with color Doppler effect
was then performed to confirm hemostasis. If necessary, an additional 1 mL of cyano-
acrylate was injected (Fig. 2).
Therapeutic success was 100% with no procedure-related complications. In follow-
up, 24 patients had repeat endoscopy by the time of study publication with complete
obliteration of gastric varices after a single session of therapy achieved in 95.8%, as
confirmed by color Doppler analysis showing no residual vascular flow. One patient
had recurrent bleeding from gastric varices at 21 days post-treatment treated suc-
cessfully with additional coil and cyanoacrylate injection. No surgical or percutaneous
shunts were required for salvage therapy. After treatment, the natural anticipated
behavior of coils is eventual extrusion into the GI lumen.
In 2016, Our centre published an expanded series involving 152 patients with
extended follow-up over a mean of 436 days.54 Among these patients, 5% received
EUS-guided glue-coil injection for active bleeding; 69% received therapy for recent
bleeding and 26% for primary prophylaxis; 143 patients had IGV1 varices; and the
other 9 had GOV2 varices. Therapeutic success was 99% using either a transgastric
or transcrural approach. In the 1 technical failure, treatment of a gastric varix with
adherent clot was performed after the patient presented with hematemesis. Persistent
bleeding was present despite treatment with a 15 mm diameter coil and injection of
6 mL of 2-octyl cyanoacrylate requiring referral for emergent TIPS.
Follow-up EUS was performed in 100 of 152 patients with a rate of complete gastric
variceal obliteration of 93% (79% single session, 10% 2 sessions, 2% 3 sessions, and
2% 4 sessions). Only 3% of patients had gastric variceal bleeding during the follow-up
period after initial obliteration was achieved. On average, bleeding occurred at
146 weeks after initial glue-coil therapy. Minor delayed bleeding related to coil

Fig. 2. (A) IGV1 varices. (B) Endosonography of same IGV1 varices. (C) EUS-guided glue coil
embolization. (D) One month post-treatment. (E) Persistent obliteration of flow through
variceal complex at 1 month. (F) Nine-month follow-up showing variceal eradication.
(From Bhat YM, Weilert F, Fredrick RT, et al. EUS-guided treatment of gastric fundal varices
with combined injection of coils and cyanoacrylate glue: a large U.S. experience over 6 years
(with video). Gastrointest Endosc 2016;83(6):1166; with permission.)
 Endoscopic Management of Portal Hypertension 331

extrusion occurred in additional 3% of patients whereas 3% had mild postprocedure

pain and 1% (1 patient) has evidence of pulmonary embolism. The pulmonary embo-
lism did not present until 1 week after treatment after having been discharged home
postprocedure with no chest pain or dyspnea and no hypoxia or other signs of acute
embolism. The rate of symptomatic embolism in this study 0.7% (1/152) was lower
than that previously reported with EUS-guided cyanoacrylate injection without coils
(2/19; 10.5%).51

Coil Injection Alone

Levy and colleagues55 reported sole injection of embolization coils without cyanoac-
rylate for therapy for ectopic variceal bleeding. Refractory bleeding from choledocho-
jejunal anastomosis varices was treated with injection of an embolization coil through
a 22-gauge needle using a linear echoendoscope. Three coils were used with obliter-
ation of flow observed after 10 minutes. Rebleeding occurred with repeat EUS per-
formed. There was persistent absence of flow in the previously treated varices but
injection of 2 additional coils in untreated varices was performed.
EUS-guided coil embolization therapy was reported for gastric varices by Romero-
Castro and colleagues56 in 2010. EUS-guided coil embolization has been compared
retrospectively to EUS-guided cyanoacrylate injection for therapy for gastric varices.51
Eleven patients were treated with coils (coiled diameter ranging from 8 mm to 20 mm),
with coil size chosen approximately 20% larger than the variceal lumen diameter.
Obliteration was achieved with coil injection in 90.9% of patients. Complications
were significantly lower in the coil group versus the cyanoacrylate injection group
(58% vs 9%; P<.01). Symptomatic pulmonary embolism occurred in 2 of 19 patients
after cyanoacrylate injection whereas only 1 of 11 patients after coil injection had an
adverse event. As discussed previously, however, CT imaging showed that 11 of
19 patients actually had evidence of pulmonary embolization after cyanoacrylate
injection.
Occasionally, the authors have noted patients suffering suspected gastric variceal
bleeding with only small diameter gastric varices present. In such cases, sole coil
embolization therapy without cyanoacrylate injection can be effective and allows
avoidance of extravascular glue injection into the submucosal space, which otherwise
can rapidly compress and obfuscate the endosonographic view of target vessels.
Typically, the authors deploy 1 or more 10-mm–diameter  7-cm–length coils for ther-
apy in this instance until obliteration of flow is achieved.

Gastric Varices Primary Prophylaxis

Although gastric variceal bleeding occurs less commonly than esophageal variceal
bleeding, the potential lethality of a gastric variceal bleed remains alarming. In exam-
ination of the role of primary prophylactic therapy, an RCT involving 89 patients with
either GOV2 varices with eradicated esophageal varices or IGV1 varices greater
than 1 cm without prior bleeding were managed with either cyanoacrylate injection,
medical intervention with nonselective b-blockers, or observation.57 Cyanoacrylate in-
jection resulted in lower rates of hemorrhage (10%) compared with either nonselective
b-blockers (38%) or observation (53%) strategies. Cyanoacrylate management group
survival was statistically equivalent to the nonselective b-blockers group (93% vs
83%) but higher compared with that of the observation group (74%).
Review of bleeding events showed that a large-size (varix >20 mm), high Model for
End-Stage Liver Disease score greater than 16, and presence of portal gastropathy all
were predictive high-risk factors for bleeding. Based on this trial, prophylactic inter-
vention should be considered when large or high-risk gastric varices are present for
332 Nett & Binmoeller

reduction of bleeding and mortality. In the series of 152 patients treated with EUS-
guided glue coil embolization, 40 received therapy for primary prophylaxis, with com-
plete obliteration rate of 96% with no procedure-related complications. Based on
these outcomes, the authors propose routine prophylactic therapy for IGV1 and
GOV2 varices larger than 2 cm in size.54

PORTAL GASTROPATHY

Limited effective endoscopic options exist for the management of portal hypertensive
gastropathy (PHG) bleeding. In PHG, ectatic mucosal capillaries and venules develop,
which may result in diffuse and recurrent bleeding presenting acutely or as chronic
occult blood loss. APC has been evaluated for treatment of PHG with suggestion
that, in combination with nonselective b-blocker therapy, it can reduce rates of blood
transfusion and ICU admission and improve hemoglobin levels.58 Spray application of
a hemostatic agent is also an option for treatment of active PHG bleeding. Although
this therapy achieves no long-term control of PHG, it may be used as an effective
method of acute hemostasis until nonselective b-blocker therapy, TIPS, or correction
of coagulopathy is pursued.59

RECTAL VARICES

Rectal varices develop from submucosal portosystemic connections between the

midrectum and anorectal junction.60 They exist between approximately 40% and
55% of patients with cirrhosis and between approximately 60% to 90% of patients
with extrahepatic portal vein hypertension.6 Bleeding occurs from rectal varices in
approximately 0.5% to 5% of patients who develop them. They are the source of
ectopic variceal bleeding in approximately 8% of cases of ectopic bleeding. Most
rectal varices are supplied by the inferior mesenteric vein or the superior rectal vein,
a branch of the inferior mesenteric vein, that feeds into the intrinsic rectal venous
plexus spanning along the lateral rectal wall. Systemic system outflow from this plexus
occurs into the middle and inferior rectal veins or the internal iliac vein. Rectal varices
should not be confused with internal hemorrhoids, which arise from submucosal arte-
riovenous connections at the anorectal vascular plexus without communication with
major vessels of the portal venous system. Rectal varices do not extend to the anal
columns and dentate line. Although frequently detected with endoscopic examination,
endosonography provides a superior means of diagnosis.61
No consensus exists for a standard endoscopic (or overall) management approach
to rectal varices. Options for endoscopic therapy include sclerosant injection, cyano-
acrylate glue injection, glue-coil embolization, and band ligation. If band ligation is per-
formed, EUS characterization first may help to direct banding by enabling localization
of inflowing vessels supplied by the superior rectal vein as they penetrate through the
rectal wall. Banding of rectal varices can then be performed at the site of these pene-
trating vessels and proceed distally toward the anorectal junction.60 Band ligation of
rectal varices has been associated with high recurrence rates, however. Comparison
with sclerotherapy has shown reduced bleeding recurrence after sclerotherapy versus
band ligation—33.3% versus 55.6%.62 For complete obliteration, sclerosant injection
into rectal varices may require use of large volumes compared with cyanoacrylate.
Thus, cyanoacrylate use, with injection of smaller volumes, may at least theoretically
reduce risk of systemic embolization. As with gastric varices, use of an EUS-guided
injection of an embolization coil followed by cyanoacrylate may be used to act as a
scaffolding for cyanoacrylate fixation for further reduction of distant embolization
risk (Fig. 3).63
 Endoscopic Management of Portal Hypertension 333

Fig. 3. (A) Endoscopic view of rectal varix Black arrow shows nipple sign at gastric varix. (B)
EUS visualization. (C) Status post–glue-coil embolization with no residual flow. (From Wei-
lert F, Shah JN, Marson FP, et al. EUS-guided coil and glue for bleeding rectal varix. Gastro-
intest Endosc 2012;76(4):915–6; with permission.)

OTHER ECTOPIC VARICES

Ectopic varices are a rare cause of bleeding in patients with portal hypertension, ac-
counting for only 2% to 5% of variceal bleeding cases.64 They may be more common
in patients with prehepatic etiologies of portal hypertension rather than cirrhosis,
occurring in 27% to 40% of patients with splanchnic vein thromboses.65 Bleeding
from ectopic varices also seems rarer but more severe than esophageal variceal
bleeding, with mortality rates of up to 40%.64,66 In a retrospective case series of
169 patients, peristomal, duodenal, and jejunoileal varices were the most common
sites of ectopic variceal bleeding.67 Biliary tract, colonic, periumbilical, and peritoneal
varices may occur as well.68
No standardized treatment exists for ectopic varices, but endoscopic treatment
options include those treatments used for varices in the stomach and esoph-
agus—band ligation, EUS-guided coil embolization, and/or cyanoacrylate injection.2
Band ligation should be pursued with caution if the diameter of the varix is too large
to fully entrap within a ligature. Hemostatic clip placement has been reported as well
for ectopic variceal therapy though failure of obliteration and rebleeding may be
anticipated.64 APC in conjunction with band ligation may be considered for preven-
tion of variceal occurrence, as has been reported in the treatment of ileocolonic
anastomotic varices.68 Endoscopic therapy may be complicated by postligation ulcer
bleeding, pyelephlebitis, and portal biliopathy. Given the overall dearth of data sup-
porting endoscopic obliteration and long-term control, the role of percutaneous
embolization and shunting and surgical interventions should always be considered
for ectopic variceal management.

SUMMARY

The management of portal hypertension related bleeding requires a multidisciplinary

approach involving hepatology, interventional radiology, surgery, and luminal and
interventional gastroenterology. For hemostasis and prophylaxis against initial and
recurrent bleeding, endoscopic therapy is a critical modality. A variety of techniques,
including band ligation, sclerosant injection, tissue adhesive injection, and coil embo-
lization, are available for endoscopic therapy. When injection therapy is used, endoso-
nography can enhance the therapeutic value of intervention.

REFERENCES

1. Bloom S, Kemp W, Lubel J. Portal hypertension: pathophysiology, diagnosis and

management. Intern Med J 2015;45(1):16–26.
334 Nett & Binmoeller

2. Garcia-Tsao G, Abraldes JG, Berzigotti A, et al. Portal hypertensive bleeding in

cirrhosis: Risk stratification, diagnosis, and management: 2016 practice guid-
ance by the American Association for the study of liver diseases. Hepatology
2017;65(1):310–35.
3. Sharma M, Rameshbabu CS. Collateral pathways in portal hypertension. J Clin
Exp Hepatol 2012;2(4):338–52.
4. Van Stiegmann G, Cambre T, Sun JH. A new endoscopic elastic band ligating de-
vice. Gastrointest Endosc 1986;32(3):230–3.
5. Poza Cordon J, Froilan Torres C, Burgos Garcı́a A, et al. Endoscopic manage-
ment of esophageal varices. World J Gastrointest Endosc 2012;4(7):312–22.
6. Philips CA, Augustine P. Endoscopic ultrasound-guided management of
bleeding rectal varices. ACG Case Rep J 2017;4:e101.
7. Vianna A, Hayes PC, Moscoso G, et al. Normal venous circulation of the gastro-
esophageal junction. A route to understanding varices. Gastroenterology 1987;
93(4):876–89.
8. Kapoor A, Dharel N, Sanyal AJ. Endoscopic diagnosis and therapy in gastro-
esophageal variceal bleeding. Gastrointest Endosc Clin N Am 2015;25(3):
491–507.
9. Nijhawan S, Rai RR, Nepalia S, et al. Natural history of postligation ulcers. Am J
Gastroenterol 1994;89(12):2281–2.
10. Cho E, Jun CH, Cho SB, et al. Endoscopic variceal ligation-induced ulcer
bleeding: what are the risk factors and treatment strategies? Medicine (Baltimore)
2017;96(24):e7157.
11. Sanglodkar UA, Jothimani D, Rela M. Hemospray for recurrent esophageal band
ulcer bleeding. Clin Exp Hepatol 2018;4(1):46–8.
12. Tierney A, Toriz BE, Mian S, et al. Interventions and outcomes of treatment of
postbanding ulcer hemorrhage after endoscopic band ligation: a single-center
case series. Gastrointest Endosc 2013;77(1):136–40.e1.
13. Ochiai T, Nakade Y, Kitano R, et al. Hemothorax following uncomplicated endo-
scopic variceal sclerotherapy and ligation for esophageal varices. Case Rep
Gastroenterol 2017;11(3):531–8.
14. Dai C, Liu WX, Jiang M, et al. Endoscopic variceal ligation compared with endo-
scopic injection sclerotherapy for treatment of esophageal variceal hemorrhage:
a meta-analysis. World J Gastroenterol 2015;21(8):2534–41.
15. Laine L, Cook D. Endoscopic ligation compared with sclerotherapy for treatment
of esophageal variceal bleeding. A meta-analysis. Ann Intern Med 1995;123(4):
280–7.
16. Karsan HA, Morton SC, Shekelle PG, et al. Combination endoscopic band ligation
and sclerotherapy compared with endoscopic band ligation alone for the sec-
ondary prophylaxis of esophageal variceal hemorrhage: a meta-analysis. Dig
Dis Sci 2005;50(2):399–406.
17. de Franchis R, Baveno VIF. Expanding consensus in portal hypertension: report
of the baveno VI consensus workshop: stratifying risk and individualizing care for
portal hypertension. J Hepatol 2015;63(3):743–52.
18. Cardenas A, Fernandez-Simon A, Escorcell A. Endoscopic band ligation and
esophageal stents for acute variceal bleeding. Clin Liver Dis 2014;18(4):793–808.
19. McCarty TR, Njei B. Self-expanding metal stents for acute refractory esophageal
variceal bleeding: a systematic review and meta-analysis. Dig Endosc 2016;
28(5):539–47.
20. Shao XD, Qi XS, Guo XZ. Esophageal stent for refractory variceal bleeding: a sys-
temic review and meta-analysis. Biomed Res Int 2016;2016:4054513.
 Endoscopic Management of Portal Hypertension 335

21. Escorsell A, Pavel O, Cárdenas A, et al. Esophageal balloon tamponade versus

esophageal stent in controlling acute refractory variceal bleeding: a multicenter
randomized, controlled trial. Hepatology 2016;63(6):1957–67.
22. Cipolletta L, Bianco MA, Rotondano G, et al. Argon plasma coagulation prevents
variceal recurrence after band ligation of esophageal varices: preliminary results
of a prospective randomized trial. Gastrointest Endosc 2002;56(4):467–71.
23. Li X, Jiang T, Gao J. Endoscopic variceal ligation combined with argon plasma
coagulation versus ligation alone for the secondary prophylaxis of variceal
bleeding: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol
2017;29(6):621–8.
24. Yol S, Belviranli M, Toprak S, et al. Endoscopic clipping versus band ligation in
the management of bleeding esophageal varices. Surg Endosc 2003;17(1):
38–42.
25. Naga MI, Okasha HH, Foda AR, et al. Detachable endoloop vs. elastic band liga-
tion for bleeding esophageal varices. Gastrointest Endosc 2004;59(7):804–9.
26. Santos MM, Tolentino LH, Rodrigues RA, et al. Endoscopic treatment of esopha-
geal varices in advanced liver disease patients: band ligation versus cyanoacry-
late injection. Eur J Gastroenterol Hepatol 2011;23(1):60–5.
27. Cipolletta L, Zambelli A, Bianco MA, et al. Acrylate glue injection for acutely
bleeding oesophageal varices: a prospective cohort study. Dig Liver Dis 2009;
41(10):729–34.
28. Ribeiro JP, Matuguma SE, Cheng S, et al. Results of treatment of esophageal var-
iceal hemorrhage with endoscopic injection of n-butyl-2-cyanoacrylate in patients
with Child-Pugh class C cirrhosis. Endosc Int Open. 3(6):E584–589.
29. Ljubicic N, Bisc anin A, Nikolic
 M, et al. A randomized-controlled trial of endo-
scopic treatment of acute esophageal variceal hemorrhage: N-butyl-2-cyanoac-
rylate injection vs. variceal ligation. Hepatogastroenterology 2011;58(106):
438–43.
30. Maluf-Filho F, Sakai P, Ishioka S, et al. Endoscopic sclerosis versus cyanoacrylate
endoscopic injection for the first episode of variceal bleeding: a prospective,
controlled, and randomized study in Child-Pugh class C patients. Endoscopy
2001;33(5):421–7.
31. Feretis C, Dimopoulos C, Benakis P, et al. N-butyl-2-cyanoacrylate (Histoacryl)
plus sclerotherapy versus sclerotherapy alone in the treatment of bleeding
esophageal varices: a randomized prospective study. Endoscopy 1995;27(5):
355–7.
32. Thakeb F, Salama Z, Salama H, et al. The value of combined use of N-butyl-2-
cyanoacrylate and ethanolamine oleate in the management of bleeding esopha-
gogastric varices. Endoscopy 1995;27(5):358–64.
33. Al-Osaimi FR, Bennamoun M, Mian A. Spatially optimized data-level fusion of
texture and shape for face recognition. IEEE Trans Image Process 2012;21(2):
859–72.
34. Sharma BC, Varakanahalli S, Singh JP, et al. Gastric varices in cirrhosis vs. extra-
hepatic portal venous obstruction and response to endoscopic N-butyl-2-cyano-
acrylate injection. J Clin Exp Hepatol 2017;7(2):97–101.
35. Binmoeller KF. Endoscopic ultrasound-guided coil and glue injection for gastric
variceal bleeding. Gastroenterol Hepatol (N Y) 2018;14(2):123–6.
36. Sarin SK, Lahoti D, Saxena SP, et al. Prevalence, classification and natural history
of gastric varices: a long-term follow-up study in 568 portal hypertension patients.
Hepatology 1992;16(6):1343–9.
336 Nett & Binmoeller

37. Trudeau W, Prindiville T. Endoscopic injection sclerosis in bleeding gastric vari-

ces. Gastrointest Endosc 1986;32(4):264–8.
38. Sarin SK. Long-term follow-up of gastric variceal sclerotherapy: an eleven-year
experience. Gastrointest Endosc 1997;46(1):8–14.
39. Weilert F, Binmoeller KF. Endoscopic management of gastric variceal bleeding.
Gastroenterol Clin North Am 2014;43(4):807–18.
40. Sarin SK, Jain AK, Jain M et al. A randomized controlled trial of cyanoacrylate
versus alcohol injection in patients with isolated fundic varices. Am J Gastroen-
terol. 97(4):1010–1015
41. Rios Castellanos E, Seron P, Gisbert JP, et al. Endoscopic injection of cyanoac-
rylate glue versus other endoscopic procedures for acute bleeding gastric vari-
ces in people with portal hypertension. Cochrane Database Syst Rev
2015;(5):CD010180.
42. Soehendra N, Nam VC, Grimm H, et al. Endoscopic obliteration of large esoph-
agogastric varices with bucrylate. Endoscopy 1986;18(1):25–6.
43. de Franchis R, Primignani M. Endoscopic treatments for portal hypertension.
Semin Liver Dis 1999;19(4):439–55.
44. Tantau M, Crisan D, Popa D, et al. Band ligation vs. N-Butyl-2-cyanoacrylate in-
jection in acute gastric variceal bleeding: a prospective follow-up study. Ann
Hepatol 2013;13(1):75–83.
45. El Amin H, Abdel Baky L, Sayed Z, et al. A randomized trial of endoscopic vari-
ceal ligation versus cyanoacrylate injection for treatment of bleeding junctional
varices. Trop Gastroenterol 2010;31(4):279–84.
46. Levy I, Binmoeller KF. EUS-guided vascular interventions. Endosc Ultrasound
2018;7(4):228–35.
47. Boustiere C, Dumas O, Jouffre C, et al. Endoscopic ultrasonography classifica-
tion of gastric varices in patients with cirrhosis. Comparison with endoscopic find-
ings. J Hepatol 1993;19(2):268–72.
48. Iwase H, Suga S, Morise K, et al. Color Doppler endoscopic ultrasonography for
the evaluation of gastric varices and endoscopic obliteration with cyanoacrylate
glue. Gastrointest Endosc 1995;41(2):150–4.
49. Romero-Castro R, Pellicer-Bautista FJ, Jimenez-Saenz M, et al. EUS-guided in-
jection of cyanoacrylate in perforating feeding veins in gastric varices: results
in 5 cases. Gastrointest Endosc 2007;66(2):402–7.
50. Battaglia G, Morbin T, Patarnello E, et al. Visceral fistula as a complication of
endoscopic treatment of esophageal and gastric varices using isobutyl-2-
cyanoacrylate: report of two cases. Gastrointest Endosc 2000;52(2):267–70.
51. Romero-Castro R, Ellrichmann M, Ortiz-Moyano C, et al. EUS-guided coil versus
cyanoacrylate therapy for the treatment of gastric varices: a multicenter study
(with videos). Gastrointest Endosc 2013;78(5):711–21.
52. Burke MP, O’Donnell C, Baber Y. Death from pulmonary embolism of cyanoacry-
late glue following gastric varix endoscopic injection. Forensic Sci Med Pathol
2017;13(1):82–5.
53. Binmoeller KF, Weilert F, Shah JN, et al. EUS-guided transesophageal treatment
of gastric fundal varices with combined coiling and cyanoacrylate glue injection
(with videos). Gastrointest Endosc 2011;74(5):1019–25.
54. Bhat YM, Weilert F, Fredrick RT, et al. EUS-guided treatment of gastric fundal vari-
ces with combined injection of coils and cyanoacrylate glue: a large U.S. expe-
rience over 6 years (with video). Gastrointest Endosc 2016;83(6):1164–72.
 Endoscopic Management of Portal Hypertension 337

55. Levy MJ, Wong Kee Song LM, Kendrick ML, et al. EUS-guided coil embolization
for refractory ectopic variceal bleeding (with videos). Gastrointest Endosc 2008;
67(3):572–4.
56. Romero-Castro R, Pellicer-Bautista F, Giovannini M, et al. Endoscopic ultrasound
(EUS)-guided coil embolization therapy in gastric varices. Endoscopy 2010;
42(Suppl 2):E35–6.
57. Mishra SR, Sharma BC, Kumar A, et al. Primary prophylaxis of gastric variceal
bleeding comparing cyanoacrylate injection and beta-blockers: a randomized
controlled trial. J Hepatol 2011;54(6):1161–7.
58. Hanafy AS, El Hawary AT. Efficacy of argon plasma coagulation in the manage-
ment of portal hypertensive gastropathy. Endosc Int Open 2016;4(10):E1057–62.
59. Smith LA, Morris AJ, Stanley AJ. The use of hemospray in portal hypertensive
bleeding; a case series. J Hepatol 2014;60(2):457–60.
60. Sharma M, Rai P, Bansal R. EUS-assisted evaluation of rectal varices before
banding. Gastroenterol Res Pract 2013;2013:619187.
61. Al Khalloufi K, Laiyemo AO. Management of rectal varices in portal hypertension.
World J Hepatol 2015;7(30):2992–8.
62. Sato T, Yamazaki K, Akaike J, et al. Retrospective analysis of endoscopic injec-
tion sclerotherapy for rectal varices compared with band ligation. Clin Exp Gas-
troenterol 2010;3:159–63.
63. Weilert F, Shah JN, Marson FP, et al. EUS-guided coil and glue for bleeding rectal
varix. Gastrointest Endosc 2012;76(4):915–6.
64. Park SW, Cho E, Jun CH, et al. Upper gastrointestinal ectopic variceal bleeding
treated with various endoscopic modalities: case reports and literature review.
Medicine (Baltimore) 2017;96(1):e5860.
65. Henry Z, Uppal D, Saad W, et al. Gastric and ectopic varices. Clin Liver Dis 2014;
18(2):371–88.
66. Orr DW, Harrison PM, Devlin J, et al. Chronic mesenteric venous thrombosis:
evaluation and determinants of survival during long-term follow-up. Clin Gastro-
enterol Hepatol 2007;5(1):80–6.
67. Norton ID, Andrews JC, Kamath PS. Management of ectopic varices. Hepatology
1998;28(4):1154–8.
68. Helmy A, Al Kahtani K, Al Fadda M. Updates in the pathogenesis, diagnosis and
management of ectopic varices. Hepatol Int 2008;2(3):322–34.