Portal hypertension in the era of endo-hepatology: Emerging diagnostic and therapeutic roles of endoscopy and endoscopic ultrasound

Abstract

Portal hypertension is a major complication of chronic liver disease and cirrhosis, causing significant morbidity and mortality through gastroesophageal varices, portal hypertensive gastropathy (PHG), and variceal hemorrhage. Although endoscopy has long been central to diagnosing and treating these complications, advances in endoscopic ultrasound (EUS) have expanded its role and given rise to the novel field of endo-hepatology, an evolving paradigm in which advanced EUS techniques transform endoscopy from a primarily therapeutic tool for variceal hemorrhage into an integrated platform for hemodynamic assessment, tissue characterization, and targeted intervention within a single session. A comprehensive narrative review of the literature was conducted to summarize both established and novel endoscopic approaches to portal hypertension management, including their indications, advantages, limitations, and integration with risk stratification tools, pharmacologic therapies, and multidisciplinary care pathways. Conventional modalities, including endoscopic variceal ligation, sclerotherapy, cyanoacrylate injection, and argon plasma coagulation, remain essential for managing esophageal and gastric varices and PHG. Newer EUS-based techniques, including EUS-guided coil and glue therapy, portal pressure gradient measurement, and liver biopsy, provide complementary approaches that enable more precise hemodynamic assessment and targeted intervention. These advances illustrate the transition from conventional bleeding-focused endoscopic management to an integrated endo-hepatology platform that supports personalized management of portal hypertension. Further studies are needed to define optimal patient selection, comparative effectiveness, and long-term outcomes.

Key Words: Portal hypertension; Cirrhosis; Esophageal varices; Gastric varices; Variceal hemorrhage; Endoscopic ultrasound; Endo-hepatology; Endoscopic variceal ligation; Portal pressure gradient

Core Tip: Endoscopy is expanding beyond its traditional hemostatic role in portal hypertension into an integrated tool that expands its clinical potential. Advances in endoscopic ultrasound-based procedures allow for a comprehensive single-session evaluation of disease severity and bleeding risk. This endo-hepatology approach facilitates risk stratification and more informed clinical decision-making and a shift toward more personalized care in patients with chronic liver disease.

INTRODUCTION

Portal hypertension is a major complication of chronic liver disease and cirrhosis, arising from increased intrahepatic vascular resistance and increased portal venous inflow[1]. Clinically significant portal hypertension (CSPH) contributes substantially to cirrhosis-related morbidity through gastroesophageal varices, ascites, hepatic encephalopathy, portal hypertensive gastropathy (PHG), and variceal hemorrhage[2,3].

Gastroesophageal varices are among the most clinically important complications, as they are common in cirrhosis and may progress to life-threatening bleeding events[4,5].

Endoscopy has long been central to the diagnosis, risk assessment, and treatment of portal hypertension-related complications. Conventional endoscopic therapies, including endoscopic variceal ligation (EVL), injection sclerotherapy, cyanoacrylate injection, and argon plasma coagulation (APC), remain essential for managing esophageal varices, gastric varices, and PHG[6]. However, conventional endoscopy is primarily a luminal and therapeutic modality, visualizing mucosal and intraluminal abnormalities and treating complications, but it does not directly measure portal pressure, characterize liver fibrosis, or fully define the collateral vascular anatomy that drives variceal formation and recurrence[6,7].

The emergence of endoscopic ultrasound (EUS) has expanded the role of endoscopy beyond conventional bleeding control. EUS allows for assessment of collateral vessels and perforating veins, targeted vascular therapy, portal pressure gradient (PPG) measurement, and EUS-guided liver biopsy[7,8]. These capabilities have contributed to the evolving field of endo-hepatology, in which endoscopy functions as an integrated platform for hemodynamic assessment, tissue characterization, and targeted intervention. This approach allows patients to undergo variceal evaluation, portal pressure assessment, liver biopsy, and therapeutic intervention during a single session.

AIM

This narrative review summarizes the evolving role of endoscopy in portal hypertension, beginning with established approaches to screening, risk stratification, and endoscopic therapy, and then focusing on emerging EUS-guided techniques that define the endo-hepatology paradigm. We also discuss indications, limitations, patient-selection considerations, and the integration of these approaches with pharmacologic therapy, noninvasive risk stratification, and multidisciplinary care. The principal endoscopic and EUS-based techniques in portal hypertension are summarized in Table 1.

Table 1 Endoscopic and endoscopic ultrasound guided techniques for the screening, diagnosis, risk stratification, and management of portal hypertension.

Technique	Mechanism	Primary indications	Advantages	Clinical outcomes	Limitations	Key evidence
Screening and risk stratification
EGD	Direct visualization and grading of varices and mucosal features (size, red wale markings, cherry red spots)	Patients with compensated cirrhosis who do not meet Baveno VI/VII criteria for deferred screening; surveillance after variceal eradication	Identifies high-risk features that predict bleeding; guides prophylaxis and surveillance	Baveno VI criteria safely defer endoscopy with only 2.2% missed high-risk varices; Baveno VII avoids 56.7%-75.4% of EGDs	Cannot assess extramural collaterals or portal hemodynamics	AASLD 2024 practice guidance (Kaplan et al[9]); Baveno VII consensus 2022 (de Franchis et al[11])
Esophageal varices
EVL	Mechanical strangulation causing ischemia, necrosis, and fibrosis	First-line for acute esophageal variceal bleeding; primary prophylaxis when NSBBs contraindicated; secondary prophylaxis (with NSBBs)	High hemostasis rate; achieves long-term variceal obliteration; well-established	Hemostasis in majority of acute bleeding cases	PBUB in 5.5% (22.5% mortality); does not reduce portal pressure; requires repeated sessions	AASLD 2024 practice guidance (Kaplan et al[9]); ESGE 2022 guideline (Gralnek et al[29]); de Brito Nunes et al[31]
EIS	Chemical sclerosant injection causing thrombosis and fibrosis	Alternative when EVL not feasible	Useful in difficult visualization during active bleeding or scarred esophagus	Effective rescue therapy	Inferior to EVL for most patients	Li et al[43]
EUS-guided sclerotherapy	Ultrasound-directed sclerosant injection targeting perforating veins	Esophageal varices with persistent perforating veins; recurrence after conventional therapy	Targets feeder vessels not visible on standard endoscopy; shorter treatment times; less severe rebleeding	Significantly shorter treatment times and greater improvement in variceal bleeding with EUS guidance	Requires EUS expertise; limited data from small studies; multicenter trials needed	Dhiman et al[32]; Fang et al[33]
Self-expanding esophageal stent	Mechanical tamponade of bleeding varices	Refractory acute variceal hemorrhage as bridge to TIPS or definitive therapy	High immediate hemostasis; avoids balloon tamponade complications	Effective temporary bleeding control	Temporary bridge; stent migration; requires removal	AASLD 2024 practice guidance (Kaplan et al[9]); ESGE guideline 2022 (Gralnek et al[29])
Gastric varices
ECI	Cyanoacrylate polymerizes on blood contact embolizing the varix	First-line for acute gastric variceal bleeding; secondary prophylaxis of gastric varices; GOV1 (ECI or EVL)	Current endoscopic standard; high hemostasis rates; low embolization rate	87%-100% hemostasis in acute bleeding	Recurrence (34%); early rebleeding (16%); late rebleeding (39%); other adverse event rate (28%)	ESGE 2022 guideline (Gralnek et al[29]); AGA clinical practice update 2021 (Henry et al[29]); Hu et al[38]
EUS-guided coil + cyanoacrylate	EUS-guided coil scaffold for thrombus formation followed by cyanoacrylate to prevent embolization	Gastric varices (acute and prophylactic)	Doppler confirmation of hemostasis; precise targeting; lower recurrence than ECI alone	84% obliteration (vs 63% direct injection); 96%-98% hemostasis; recurrence 5%; > 95% obliteration in prophylaxis	Requires advanced EUS training; limited to specialized centers; pooled adverse event rate 14%; not suitable with complex vascular anatomy	Florencio de Mesquita et al[41]; Kouanda et al[42]
Portal hypertensive gastropathy
APC	Non-contact thermal coagulation via high-frequency current through argon gas	Bleeding from PHG	Non-contact; effective alone or synergistic with NSBBs	75.4% clinical success with one session for severe PHG	May require multiple sessions; not studied in hepatic encephalopathy, severe cardiovascular disease, chronic renal failure, hemodynamic instability, or active variceal/ulcer bleeding	Hanafy and El Hawary[58]
EUS-guided portal pressure measurement
EUS-PPG	Direct measurement of portal and hepatic vein pressures via needle with compact manometer	Risk stratification; suspected presinusoidal portal hypertension; discordant noninvasive data; concurrent with variceal screening	Direct measurement (vs indirect HVPG); no radiation/contrast; same-session with endoscopy and liver biopsy	100% technical success (pilot); HVPG correlation on r = 0.923	High-risk bleeding procedure per ASGE (caution with anticoagulation); requires advanced EUS training; not yet validated in large multicenter RCTs	Huang et al[68]; Choi et al[69]; Zhang et al[70]; Delphi consensus 2025 (Wang et al[73])
EUS-guided liver biopsy
EUS-LB	Core tissue sampling under EUS visualization	Fibrosis staging; concurrent portal pressure assessment; contraindication to percutaneous biopsy	Fibrosis staging when concurrent endoscopic procedure indicated; percutaneous biopsy contraindicated (ascites, obesity, coagulopathy)	83% combined adequate biopsy + reliable pressure gradient	High-risk bleeding procedure (ASGE/AGA); requires anticoagulant interruption; caution with platelets < 50000/μL; percutaneous remains first-line when sole indication is histology	Benmassaoud et al[74]; Arruda do Espirito Santo et al[75]
Emerging EUS-guided vascular interventions
EUS-guided SPSS obliteration	EUS-directed transgastric coil and glue injection to occlude SPSS	Refractory hepatic encephalopathy with SPSS ≥ 8 mm; MELD usually < 15	Transgastric access; can combine with variceal assessment and portal pressure measurement	Clinical improvement in 6/7 patients	Avoid in advanced liver disease, portal vein thrombosis, end-stage shunt syndrome; 43% long-term adverse event rate	Rathi et al[83]; ACG 2026 guideline on HE (Bajaj et al[81]); Wang et al[85]
Hybrid EUS-TIPS	EUS localizes portal vein to guide subsequent transjugular TIPS placement	Cavernous transformation of portal vein where conventional TIPS is technically difficult	Enables TIPS in otherwise technically impossible cases	All patients successful; no recurrent hemorrhage or ascites at mean 12-month follow-up	Adds procedural complexity (requires advanced endoscopy and IR); should not be used when standard transjugular access is feasible	Zhang et al[88]
EIPS	Self-expanding metal stent via EUS bridging hepatic and portal veins	Experimental; potential alternative to TIPS	Avoids transjugular access; concurrent portal pressure measurement possible	Successful creation in porcine models with confirmed portosystemic flow	Preclinical only; classified as experimental by AGA; no human trials	Buscaglia et al[87]; AGA clinical practice update 2023 (Ryou et al[39])

This table summarizes established and emerging endoscopic and endoscopic ultrasound based techniques used in the diagnosis and management of portal hypertension, including their mechanisms, primary indications, reported efficacy, and key limitations. AASLD: American Association for the Study of Liver Diseases; AGA: American Gastroenterological Association; APC: Argon plasma coagulation; ASGE: American Society for Gastrointestinal Endoscopy; ECI: Endoscopic cyanoacrylate injection; EGD: Esophagogastroduodenoscopy; EIPS: Endoscopic intrahepatic portosystemic shunt; EIS: Endoscopic injection sclerotherapy; ESGE: European Society of Gastrointestinal Endoscopy; EUS: Endoscopic ultrasound; EUS-LB: Endoscopic ultrasound-guided liver biopsy; EUS-PPG: Endoscopic ultrasound-guided portal pressure gradient measurement; EVL: Endoscopic variceal ligation; GOV1: Gastroesophageal varices type 1; HE: Hepatic encephalopathy; HVPG: Hepatic venous pressure gradient; MELD: Model for end-stage liver disease; NSBB: Nonselective β-blocker; PHG: Portal hypertensive gastropathy; PBUB: Post-banding ulcer bleeding; SPSS: Spontaneous portosystemic shunt; TIPS: Transjugular intrahepatic portosystemic shunt; RCTs: Randomized controlled trials.

ROLE OF ENDOSCOPY IN SCREENING AND RISK STRATIFICATION

Endoscopy plays a central role in risk stratification for portal hypertension, both by identifying patients with varices and by grading features that predict bleeding risk. Historically, nearly all patients with cirrhosis underwent screening endoscopy; however, advances in noninvasive assessment have fundamentally changed this approach. Current guidelines incorporate liver stiffness measurement (LSM) and laboratory parameters to identify patients at low risk for CSPH, allowing many patients to safely defer endoscopic evaluation while reserving endoscopy for those most likely to benefit[9,10].

Determining who needs endoscopy

While liver biopsy and hepatic venous pressure gradient (HVPG) measurement remain the gold standards for diagnosing cirrhosis and CSPH, respectively, noninvasive tools have transformed patient selection for endoscopy. Current AASLD and Beveno VII guidance recommend LSM combined with platelet count to identify compensated advanced chronic liver disease and estimate the likelihood of CSPH risk[9-11].

Both guidelines use LSM and platelet count as the primary noninvasive triage tools to assess CSPH. CSPH is ruled out when LSM is < 15 kPa with platelets ≥ 150 × 10⁹/L, whereas higher LSM thresholds increased the probability. To rule in CSPH, the AASLD uses tiered LSM-platelet combinations, while Baveno VII applies LSM ≥ 25 kPa specifically to viral hepatitis, alcohol-associated liver disease, and MASLD with body mass index < 30[9,12,13]. Notably, there is no role for LSM after decompensation.

These noninvasive criteria substantially reduced unnecessary screening endoscopy. The Baveno VI criteria (LSM < 20 kPa plus platelet count > 150000/mm³) can defer endoscopy in patients at low risk for high-risk varices[13]. The Baveno VII algorithm extends this guideline by recommending endoscopy for patients that are not candidates for nonselective β-blockers (NSBBs) with LSM ≥ 20 kPa or platelets ≤ 150 × 10⁹/L[9,11]. The Baveno VII algorithm further incorporates spleen stiffness measurement (≤ 40 kPa), allowing 56.7%-75.4% of screening endoscopies to be avoided while maintaining a missed high-risk variceal rate of only 3%-3.8%[14]. Although spleen stiffness measurement correlates with HVPG, it has not consistently shown superiority over LSM and has not been validated for routine clinical practice[15,16].

Overall, LSM and platelet count remain the best validated noninvasive tools to risk-stratify patients with compensated cirrhosis and determine the need for endoscopic evaluation[9]. Once endoscopy is indicated, direct visualization of variceal morphology provides additional prognostic information that guides surveillance and management decisions.

Endoscopic risk assessment

When endoscopy is indicated, grading of variceal features provides important prognostic information that directly guides therapeutic decisions. High-risk features include variceal size > 5 mm, red wale markings, and cherry red spots[17,18], all of which correlate with an increased risk of first variceal bleeding[19]. The Italian Liver Cirrhosis Project classification similarly found that features such as variceal size and congestive gastropathy independently predict the first bleeding event[20].

Beyond standard endoscopic grading, EUS provides additional risk stratification by characterizing the paraoesophageal collateral venous system, which is not visible on conventional endoscopy. EUS can detect the small periesophageal collateral veins (small vessels adjacent to the esophageal wall), para-esophageal collateral veins (larger extrinsic vessels), and perforating veins that connect the two systems[21]. These endoscopic findings have strong predictive value for variceal recurrence and rebleeding after endoscopic therapy. A systematic review found that perforating veins and peri-esophageal collateral veins are significantly associated with higher variceal recurrence within 1 year after endoscopic treatment[22]. Additionally, Carneiro et al[23] found in a prospective study that paraesophageal variceal diameter measured by EUS predicted variceal recurrence within 1 year after EVL eradication, while Masalaite et al[24] found that severe and multiple peri-esophageal collateral veins on EUS were independent predictors of variceal recurrence after EVL. Additionally, Doppler EUS assessment of azygos blood flow correlates with portal hypertension severity and permits evaluation of the hemodynamic response to vasoactive agents such as terlipressin and somatostatin[25,26]. These EUS-based collateral assessments complement standard endoscopic grading and may help identify patients at the highest risk for variceal recurrence who warrant closer surveillance or alternative therapeutic strategies.

THERAPEUTIC ENDOSCOPY IN PORTAL HYPERTENSION

Esophageal varices

Esophageal varices remain the most common indication for therapeutic endoscopy in patients with portal hypertension[27]. For primary prophylaxis, the 2024 AASLD Practice Guidance recommends NSBBs, preferably carvedilol, as first-line therapy for patients with compensated cirrhosis and high-risk varices as they reduce portal pressure and improve survival[9,10,28]. EVL is recommended for patients who have contraindications to, intolerance of, or inadequate response to NSBBs, and is the first-line endoscopic therapy during acute variceal hemorrhage[9,29,30]. The established and emerging endoscopic techniques discussed throughout this section and the remainder of the review are summarized in Table 1.

During acute variceal bleeding, EVL should be performed within 12 hours following hemodynamic resuscitation and should be repeated every 2-4 weeks until complete variceal eradication[9,29,30]. Balloon tamponade and self-expanding esophageal stents can serve as temporary rescue therapies when bleeding cannot be immediately controlled, while prophylactic transjugular intrahepatic portosystemic shunt (TIPS) placement within 72 hours is recommended for carefully selected patients at high risk of treatment failure[9,10,29,30]. Following recovery from the initial hemorrhage, combination therapy with NSBBs and serial EVL is recommended for secondary prophylaxis within 7 days, as this technique reduces recurrent bleeding more effectively than either treatment alone[9,29].

Although EVL is highly effective for achieving hemostasis and long-term variceal eradication, treatment is directed at the luminal varix rather than the portal pressure that contributes to recurrent disease[9,29]. In addition, EVL requires repeated treatment sessions and carries a risk of post-banding ulcer bleeding, which occurs in approximately 5.5% of procedures and is associated with substantial mortality[31]. These limitations have prompted investigation into EUS-guided therapeutic approaches that target the underlying collateral vessels responsible for persistent and recurrent varices.

Dhiman et al[32] found that persistent perforating veins were present in 90% of patients who failed conventional sclerotherapy compared with none of the responders, establishing these vessels as an important mechanism of treatment failure. More recently, Fang et al[33] reported that EUS-guided sclerotherapy resulted in shorter procedure times, greater improvement in variceal bleeding, reduced bleeding severity, and longer readmission-free intervals than conventional sclerotherapy. These findings suggest that targeting feeder vessels rather than the luminal varix alone may improve treatment durability, although larger multicenter studies are required before routine implementation.

Research on the use of EUS-guided coil embolization for isolated esophageal varices is limited. Fujii-Lau et al[34] found that EUS-guided coil embolization was only technically feasible in patients who had failed conventional therapy; additionally, only one patient in the study had isolated esophageal varices, limiting conclusions regarding efficacy in this population. At present, EUS-guided vascular therapies for esophageal varices should be considered promising adjunctive techniques for selected refractory cases at experienced centers. The broader diagnostic and therapeutic applications of EUS within the evolving field of endo-hepatology are discussed in later sections.

Gastric varices

Although less common than esophageal varices, gastric varices are associated with more severe hemorrhage, higher transfusion requirement, and greater mortality. The Sarin classification categorizes gastric varices into gastroesophageal varices (GOV1 and GOV2) and isolated gastric varices (IGV1 and IGV2), with fundal varices (GOV2 and IGV1) carrying the highest risk of severe bleeding. Due to their larger size, deeper submucosal location, and higher blood flow, gastric varices require therapeutic approaches distinct from those used for esophageal varices[35,36].

Endoscopic cyanoacrylate injection (ECI) remains the standard endoscopic treatment for acute gastric variceal hemorrhage, with hemostasis rates of 87%-100%[9,37]. The 2022 ESGE guideline recommends ECI as first-line therapy for bleeding gastric varices, whereas either ECI or EVL may be considered for GOV1 varices[29]. For secondary prophylaxis, repeat cyanoacrylate injection every 2-4 weeks until variceal obliteration, followed by surveillance endoscopy, remains the current standard[9].

Despite its effectiveness, conventional ECI has several limitations. Conventional direct endoscopic injection is technically complex, with risks including loss of visualization during endoscope retroflexion, inaccurate needle placement, and increased rebleeding post-intervention[38,39]. Additionally, recurrence and rebleeding remain clinical concerns despite successful initial hemostasis. A systematic review reported pooled gastric variceal recurrence, early rebleeding, and late rebleeding rates of 34%, 16%, and 39%, respectively, following ECI[38]. Although clinically significant glue embolization is uncommon[30], complications including pulmonary embolism, portal vein thrombosis, systemic embolization, infection, and needle impaction have all been reported[38]. Patients with portal vein thrombosis, ascites, active bleeding, or large paragastric collateral veins appear to be at greatest risk for treatment failure and recurrent bleeding[38].

These limitations have driven the development of EUS-guided combination therapy with coil embolization and cyanoacrylate injection, which offers better efficacy and durability compared to conventional techniques to manage gastric varices. EUS allows for accurate injection precision even during active bleeding, enables Doppler access to confirm establishment of hemostasis, and ensures that treatment is targeted correctly to intramural and perforator vessels[39]. EUS guidance also allows for the placement of hemostatic coils into the gastric varix so that thrombus can form. Cyanoacrylate is then injected to prevent adverse embolic consequences[39]. Multiple studies have found that EUS-guided coil plus glue is superior to direct injection alone, with 84% gastric variceal obliteration (vs 63% with direct endoscopic injection), a recurrence rate of 5%, and hemostasis of 96%-98%[40,41]. Furthermore, primary prophylaxis of high-risk gastric varices, defined as > 10 mm or with cherry red spots, EUS-guided coil plus cyanoacrylate injection resulted in excellent findings, with over 95% variceal obliteration with less than 5% adverse events over 3 years of follow-up[42].

Beyond coil-assisted therapy, additional EUS-guided therapeutic techniques continue to emerge. Recent studies found that EUS-guided coil embolization plus cyanoacrylate needed less reintervention when compared to cyanoacrylate injection alone (20.8% vs 53.8%)[41,42]. Less commonly, combination therapy with endoscopic clip assistance has been studied in comparison to coil embolization. A retrospective study by Li et al[43] found no significant differences in gastric variceal rebleeding rate, technical success rate, other causes of rebleeding rate, other adverse events rate, and mortality between combination therapy for secondary prophylaxis with clipping and cyanoacrylate vs coil and cyanoacrylate.

Another relatively understudied procedure is the EUS approach to sclerosant injection. Miyazaki et al[44] studied the efficacy and long-term outcomes of sclerotherapy, particularly n-butyl-2-cyanoacrylate and ethanolamine oleate, finding that conventional injection sclerotherapy can achieve durable obliteration of gastric varices, with patients who achieved complete variceal disappearance remaining free of rebleeding after the 51-month follow-up period. Literature for EUS-guided sclerotherapy for gastric varices is overall limited.

Despite these findings, EUS-guided therapy has important limitations. A 2020 meta-analysis reported a pooled overall adverse event rate of 14% with EUS-guided variceal therapy, with a significant difference between cyanoacrylate injection alone (21%) and cyanoacrylate with coiling (10%)[45]. The technique requires advanced endoscopic training and EUS equipment, limiting its availability to specialized centers[39]. Additionally, EUS-guided therapy may not be suitable when cross-sectional imaging reveals complex vascular anatomy, favoring endovascular approaches[45].

Despite the promising results of EUS-guided therapy, endoscopic treatment is not appropriate for every patient. Selection of therapy should be individualized according to vascular anatomy, portal hemodynamics, and patient factors. Consequently, management of gastric varices often requires collaboration with interventional radiology, as endovascular therapies serve as important complementary or alternative treatment options in selected patients. Balloon-occluded retrograde transvenous obliteration (BRTO) is a procedure used in the treatment of gastric varices in which gastrorenal shunt outflow is blocked through balloon catheter insertion and inflation, along with subsequent sclerosant injection[46]. One meta-analysis of 1016 patients found a high (97.3%) clinical success rate (no recurrence or rebleed of gastric varices, or complete obliteration of varices on subsequent imaging) in the use of BRTO[47]. Plug-assisted retrograde transvenous obliteration (PARTO) is an updated version of BRTO, during which a vascular plug is utilized, in place of balloon occlusion, with subsequent gelatin sponge injection[48]. Comparison of BRTO vs PARTO in a study by Cho et al[49] found significantly higher rates of complete obliteration in BRTO (100%) than in PARTO group (80%). Further, there was a higher frequency of post-embolization syndrome in those who underwent PARTO[39]. Both have cited complications of increased ascites and worsening of esophageal varices due to increased portal venous pressure[48]. BRTO has a role in recurrent gastric variceal hemorrhage prevention in a similar context to TIPS and can be considered when there is a contraindication to TIPS in patients with established gastrorenal shunt[50,51].

Overall, management of gastric varices has evolved beyond conventional luminal therapy. While ECI remains the current standard of care, EUS-guided therapies offer greater procedural precision and may improve long-term durability by directly treating the collateral circulation responsible for recurrent disease. These endoscopic advances should be considered within a broader multidisciplinary framework that also includes established endovascular techniques such as BRTO and PARTO for appropriately selected patients.

ENDOSCOPY BEYOND ESOPHAGEAL AND GASTRIC VARICES

PHG

PHG is a common manifestation of portal hypertension characterized by gastric mucosal vascular ectasia that develops secondary to portal hypertension[52,53]. Its prevalence in patients with cirrhosis ranges from 20%-75% and increases with worsening portal hypertension and hepatic decompensation[54]. Although many patients remain asymptomatic, PHG is associated with more severe portal hypertension and is often identified during active hemorrhage during endoscopy[53].

Endoscopically, PHG is characterized by a mosaic or “snakeskin” mucosal pattern, with or without red spots, most commonly involving the gastric fundus and body[55]. Several endoscopic classification systems, including those proposed by Gjeorgjievski and Cappell[56], McCormack et al[57], and the New Italian Endoscopic Club, categorize PHG as mild or severe based on these characteristic findings. Severe PHG is associated with an increased risk of bleeding and therefore has important therapeutic implications. Because PHG may coexist with portal hypertensive coagulopathy and other causes of gastrointestinal bleeding, careful endoscopic evaluation remains essential to identify the bleeding source and guide management[53].

Management of PHG aims to reduce portal pressure while controlling active recurrent bleeding[53]. NSBBs remain the first-line therapy for patients with severe PHG without varices or mild PHG with small esophageal varices and have also been shown to reduce chronic bleeding associated with PHG[53,58]. Additionally, endoscopic administration of APC combined with NSBBs is also an effective way to control severe PHG bleeding[58]. APC is a noncontact electrosurgical endoscopic technique that delivers high-frequency current through ionized argon gas to achieve superficial thermal coagulation of the bleeding mucosa[58]. In patients with severe PHG, APC achieved clinical success in 75.4% of patients after a single treatment session, while an additional 8.6% of patients with diffuse pangastric disease required multiple treatment sessions[58]. APC may be used in combination with NSBBs to provide synergistic control of bleeding or can be used alone as an effective alternative when NSBBs are contraindicated[58].

Despite its efficacy, APC has some limitations. Current evidence is derived from smaller studies, repeat treatment sessions are frequently required in patients with extensive disease, and its use has not been adequately evaluated in patients with acute hepatic encephalopathy, severe cardiovascular disease, chronic renal failure, hemodynamic instability, or active bleeding from varices or peptic ulcers[58]. Overall, PHG provides an example of the expanding role of endoscopy in portal hypertension management beyond variceal treatment.

Ectopic varices

Ectopic varices are portosystemic collateral veins that develop outside the gastroesophageal region and represent an uncommon but potentially life-threatening complication of portal hypertension[59]. Although they account for only approximately 5% of all variceal hemorrhages, bleeding is often severe, difficult to diagnose, and associated with mortality rates approaching 40%[59,60]. Ectopic varices may occur throughout the gastrointestinal tract, including the duodenum, small intestine, colon, rectum, and peritoneum, and should be suspected in patients with portal hypertension who present with gastrointestinal bleeding despite a negative esophagogastroduodenoscopy[60]. Unlike esophageal varices, ectopic varices may rupture at lower portal pressures since bleeding risk is influenced by factors such as variceal diameter and wall tension in addition to portal pressure[60].

Endoscopy plays an important role in both the diagnosis and treatment of ectopic varices. Once identified, endoscopic therapy is considered first-line treatment for endoscopically accessible bleeding lesions in combination with standard pharmacologic management[60]. Conventional endoscopic therapy, including EVL, endoscopic injection sclerotherapy (EIS), and ECI have all been employed depending on the location of the varix[59]. For rectal varices, both EVL and EIS have had good safety profiles, although EVL is associated with higher recurrence rates than EIS[59]. Duodenal varices present a greater therapeutic challenge because of their deep collateral anatomy and propensity for massive hemorrhage. Although EVL and EIS can achieve initial hemostasis, both are associated with substantial rebleeding rates, whereas ECI, particularly N-butyl-2-cyanoacrylate, were found to promote hemostasis and have superior efficacy in the treatment of bleeding duodenal varices[59]. Challenges associated with ectopic variceal band ligation include poor visualization, difficulty isolating the bleeding varix, and difficulty engaging the banding cap, all of which may contribute to the estimated rebleeding rate of 40%[59,60].

For patients with bleeding inaccessible or refractory to endoscopic therapy, endovascular intervention can offer effective treatment. TIPS effectively reduces the portosystemic pressure gradient and is recommended for appropriately selected patients with patent portal and mesenteric venous circulation[61]. However, rebleeding remains a concern following TIPS placement for ectopic varices, with rates of 23% at one year, 26% at three years, and 32% at five years[61]. Combining TIPS alongside variceal embolization has been found to reduce rebleeding rates, suggesting TIPS can be used endovascularly concomitantly with variceal embolization for varices that cannot be managed with endoscopic therapy[61].

NEW ENDOSCOPIC HEMOSTATIC TOOLS

Although conventional endoscopic therapies such as EVL, ECI, APC, and injection therapy remain the cornerstone of managing portal hypertensive bleeding, newer endoscopic hemostatic technologies have expanded treatment options for refractory or challenging hemorrhage. Among these, hemostatic powders and over-the-scope clip (OTSC) systems have shown promising results as adjunctive therapies when conventional endoscopic techniques are not feasible.

Hemospray is an inorganic hemostatic powder delivered through the working channel of an endoscope that forms a mechanical barrier upon contact with moisture, rapidly sealing the bleeding site without tissue absorption or metabolism[60]. The hemostatic powders are delivered to the bleeding site via a catheter placed through the channel of the endoscope. Since application requires less procedural precision than conventional injection or clipping techniques, hemospray may be particularly useful for diffuse bleeding, poorly accessible lesions, or patients who are not responding to standard endoscopic therapy[62]. In emergency endoscopy, hemospray was found to have immediate hemostasis rates of 97%, although long-term clinical success decreases to 65.7% at 30 days as rebleeding remains common (31%), indicating that hemospray provides an efficacious means of controlling acute gastrointestinal bleeds[63]. The noncontact, rapidly procoagulative nature of hemospray also makes it a feasible option for patients with reduced coagulation or taking anticoagulation drugs, in whom conventional injection or clips may increase tissue injury or procedural difficulty. Overall, hemospray can be used as a bridging therapy that efficiently provides rapid hemorrhage control while definitive treatment is planned[64]. Although generally safe, rare cases of venous embolization related to high-pressure powder delivery have been reported[64].

The OTSC system represents another endoscopic technique for refractory gastrointestinal hemorrhage. OTSC uses a nitinol clip mounted on the tip of the endoscope to achieve durable mechanical compression of the bleeding source[65]. OTSC has had an overall clinical success rate of 78.4%, with success rates approaching 86% for gastrointestinal bleeding while maintaining a low overall adverse event rate of 2.1%[62]. Reported adverse events associated with endoscopic OTSC include incorrect placement site, perforation, bleeding, and infection[65]. Within portal hypertension, OTSC has primarily been described as a rescue therapy following failed EVL for acute variceal bleeding. Case reports and small series suggest that OTSC can achieve successful hemostasis in post-EVL rescue therapy and can reduce the need for surgical interventions[66]. Therefore, OTSC is a pertinent option for patients with contraindications to surgery who tolerate endoscopy well[65]. However, current evidence remains limited, and larger prospective studies are needed before OTSC can be recommended as a first-line therapy for acute variceal hemorrhage[66]. In addition, OTSC deployment may not be feasible in patients with significant esophageal, gastric, or pyloric strictures that prevent passage of the mounted device[67].

THE RISE OF ENDO-HEPATOLOGY

Portal pressure measurement: Development and validation of EUS-guided measurement

For decades, HVPG measurement has served as the reference standard for assessing portal hypertension. However, HVPG is an indirect measurement obtained through a transjugular approach requiring fluoroscopy and interventional radiology expertise, and it may underestimate portal pressure in presinusoidal portal hypertension since it measures wedged hepatic venous pressure rather than portal venous pressure directly[9].

The development of EUS-guided PPG (EUS-PPG) represents one of the most significant advances in endo-hepatology and portal hypertension assessment. The first human EUS-PPG was performed in 2014[9]. The subsequent pilot study by Huang et al[68] in 28 patients with suspected cirrhosis found 100% technical success and no major adverse effects, with PPG values correlating excellently with clinical parameters of portal hypertension, including the presence of varices, portal hypertensive gastropathy, thrombocytopenia, and strong clinical evidence of cirrhosis. This proof-of-concept study established that portal hemodynamics could be assessed safely during endoscopy and laid the foundation for future clinical investigation. Choi et al[69] expanded this study to 83 patients over 6 years, further confirming that EUS-PPG correlated well with clinical markers of portal hypertension, finding that patients with esophageal or gastric varices have significantly higher PPG with excellent technical success and no major adverse events.

Since these initial reports, growing evidence has further validated the clinical performance of EUS-PPG. Zhang et al[70] found a Pearson correlation coefficient of 0.923 (P < 0.001) between EUS-PPG and transjugular HVPG/PPG measurements in patients with acute or subacute portal hypertension, with nearly identical mean portal pressure gradients between the two techniques. More recently, Choi et al[69] enrolled 385 patients across eight centers and found that patients with esophageal varices, portal hypertensive gastropathy, and thrombocytopenia had significantly higher median PPGs EUS-PPG (P < 0.001 for each). These findings have been reinforced by higher-level evidence. A 2024 systematic review and meta-analysis by Dhindsa et al[71], including eight cohort studies (178 patients), reported pooled technical and clinical success rates of 94.6% and 85.4%, respectively, with an overall adverse event rate of 10.9%, the majority of which were mild. The most common adverse events were abdominal pain (11%) and bleeding (3.6%), with no reported perforations or procedure-related deaths, further supporting the safety and feasibility of EUS-PPG in appropriately selected patients. Similarly, a recent meta-analysis of six prospective studies with paired EUS-PPG and HVPG measurements found strong pooled correlations between the two techniques (r = 0.82; 95%CI: 0.72-0.89), further validating EUS-PPG as an accurate method for portal pressure assessment despite some variability in individual paired measurements[72].

Beyond diagnostic accuracy, EUS-PPG offers several practical advantages over HVPG. A 2025 international Delphi consensus found that clinical indications for EUS-PPG encompass all indications for HVPG and may be preferred in suspected presinusoidal portal hypertension[73]. EUS-PPG is best suited for patients with discordant noninvasive data (e.g., isolated thrombocytopenia or splenomegaly without other signs of portal hypertension), suspected presinusoidal portal hypertension where HVPG may be inaccurate, or when another indication for endoscopy is already present, such as variceal screening[39,73]. Additionally, EUS-PPG allows for portal pressure to be measured during the same session as diagnostic endoscopy and EUS-guided liver biopsy, allowing for comprehensive portal hypertension evaluation while avoiding repeated procedures and multiple exposures to radiation and contrast[39,69]. However, EUS-PPG should be used with caution in patients on anticoagulation or antithrombotic therapy, as EUS with fine-needle access is considered high-risk for bleeding by the American Society for Gastrointestinal Endoscopy, and the decision to proceed requires careful multidisciplinary review weighing procedural indication against thrombotic risk[45]. Overall, EUS-PPG offers distinct advantages over traditional HVPG for portal hypertension assessment by providing direct (rather than indirect) portal pressure measurement, avoiding potential misdiagnosis of presinusoidal portal hypertension.

EUS-guided liver biopsy

Accurate fibrosis staging is essential for portal hypertension management, as it directly informs decisions regarding NSBB initiation, TIPS candidacy, and transplant evaluation. EUS-guided liver biopsy (EUS-LB) enables this assessment during the same session as variceal screening and portal pressure measurement, making it a key component of the endo-hepatology approach[69,74].

Recent evidence has found that EUS-LB provides higher diagnostic adequacy than percutaneous and transjugular approaches, particularly when combined with portal pressure assessment. A 2025 randomized trial directly comparing a combination of EUS-LB with PPG to transjugular HVPG-LB found that combined adequate liver biopsy and reliable pressure gradient was achieved in 83% of EUS cases vs only 41% of transjugular cases[74]. Furthermore, a meta-analysis of four randomized controlled trials found significantly lower post-procedure pain scores with EUS-LB compared to percutaneous biopsy[75]. Collectively, these data suggest that EUS-LB is particularly advantageous when portal pressure measurement is also indicated, offering both higher combined diagnostic success and less post-procedure pain compared to alternative biopsy techniques.

Beyond its diagnostic performance, EUS-LB offers several advantages. Under direct ultrasonographic visualization, both hepatic lobes can be sampled, reducing potential sampling error associated with heterogeneous fibrosis distribution. In addition, EUS-LB can be performed in patients with ascites, obesity, and coagulopathy, in whom percutaneous biopsy may be challenging and dangerous, while simultaneously enabling portal pressure assessment[76,77]. Therefore, EUS-LB is best indicated when liver biopsy is needed concurrently with another endoscopic procedure or when percutaneous biopsy is contraindicated.

Despite these advantages, important limitations remain. EUS-LB is classified as a high-risk procedure by both the ASGE and American Gastroenterological Association (AGA), requiring careful peri-procedural management of anticoagulant and antithrombotic medications. Although EUS-LB has a favorable safety profile in cirrhotic patients with coagulopathy and thrombocytopenia, advanced portal hypertension and acute-on-chronic liver failure are independent risk factors for procedural bleeding, and careful preprocedural planning is recommended in these populations, particularly when platelets are < 50000/μL or fibrinogen is low[78,79]. When the sole indication is liver histology without a concurrent endoscopic need, percutaneous biopsy remains the standard first-line approach[79,80].

Emerging EUS-guided vascular interventions

Beyond diagnostic applications, EUS has emerged as a platform for novel vascular interventions in portal hypertension. One important application is the identification and treatment of spontaneous portosystemic shunts (SPSS), which develop in up to 60% of patients with cirrhosis and are associated with clinical decompensation and recurrent hepatic encephalopathy[81]. However, EUS has a limited field of view restricted to structures adjacent to the gastrointestinal lumen and cannot comprehensively map the full portosystemic collateral circulation, including splenorenal or gastrorenal shunts. Consequently, contrast-enhanced computed tomography or magnetic resonance imaging remains the standard for complete vascular mapping, and EUS should be considered complementary rather than a replacement for cross-sectional imaging[9,82].

The same EUS-guided coil-and-glue platform used for gastric variceal therapy has been extended to therapeutic shunt obliteration. Spontaneous portosystemic shunts sustain refractory hepatic encephalopathy through a “flow steal” phenomenon, whereby portal blood bypasses hepatic detoxification, reducing clearance of neurotoxins[81]. The 2026 American College of Gastroenterology Clinical Guideline on Hepatic Encephalopathy recommends evaluation for SPSS in patients with refractory hepatic encephalopathy, and consideration of shunt obliteration in select patients, particularly those with model for end-stage liver disease scores below 15 and shunts measuring ≥ 8 mm in diameter. Rathi et al[83] reported successful EUS-guided transgastric shunt obliteration using coils and glue in seven patients across nine sessions, achieving complete Doppler flow cessation in six patients and clinical improvement in six patients without procedure-related severe adverse events. EUS-guided shunt obliteration should be avoided in patients with advanced liver disease, portal vein thrombosis, or end-stage portosystemic shunt syndrome, as abrupt increases in portal pressure following shunt occlusion can precipitate life-threatening complications, including variceal bleeding, intractable ascites, and hepatorenal syndrome. A meta-analysis found a long-term adverse event rate of 43%, with ascites and variceal progression as the predominant complications[84,85]. While larger studies are still needed, EUS-guided shunt obliteration shows promise in offering advantages in accessibility and the opportunity to combine shunt occlusion with concurrent variceal assessment and portal pressure measurement during the same procedure.

Additional EUS-guided vascular interventions remain investigational but also illustrate the expanding opportunities of endo-hepatology in portal hypertensive management. Preclinical studies have shown the feasibility of creating a self-expanding metal stent via EUS to bridge the hepatic and portal veins, functioning as a portosystemic shunt analogous to TIPS[86,87]. Buscaglia et al[87] successfully created EUS-guided intrahepatic portosystemic shunts (EIPS) in 10 porcine models with confirmed portosystemic flow and no evidence of bleeding or organ damage. Schulman et al[86] similarly created EIPS with concurrent direct portal pressure measurement using lumen-apposing metal stents. Nevertheless, the AGA currently classifies EIPS as an experimental intervention, noting that clinical translation requires identification of clinical scenarios where an EUS approach proves advantageous over an interventional radiology approach[39].

While EIPS remains experimental, hybrid endoscopic-radiologic approaches have already entered early clinical practice. Zhang et al[88] found that EUS can be used to localize the portal vein before subsequent TIPS placement in patients with cavernous transformation of the portal vein, a scenario where conventional TIPS is difficult. All patients successfully underwent EUS-guided localization and TIPS with no recurrent variceal hemorrhage or ascites during a mean 12-month follow-up. Although this hybrid approach may increase procedural complexity by requiring both advanced endoscopy and interventional radiology expertise, it shows how EUS can facilitate portal venous access in anatomically challenging cases where standard transjugular techniques are difficult.

Finally, EUS-guided portal vein access has been well established in the literature, and this access platform may eventually facilitate additional vascular interventions, including portal vein stenting for portal vein thrombosis or stenosis[89,90]. However, EUS-guided portal vein stenting for the treatment of portal hypertension remains investigational, and future studies are needed to establish dedicated devices, procedural safety, and long-term clinical efficacy before routine clinical implementation.

CONCLUSION

Endoscopy remains fundamental to the management of portal hypertension, with an evolving role that now extends far beyond the diagnosis and treatment of variceal hemorrhage. Advances in noninvasive risk stratification have allowed for more selective use of endoscopy, while established endoscopic therapies continue to provide effective treatment for variceal bleeding and other complications of portal hypertension. At the same time, EUS has transformed endoscopy into a platform capable of integrating hemodynamic assessment, tissue biopsy, and targeted vascular intervention. Collectively, these developments reflect a shift from a reactive approach focused on controlling hemorrhage toward a more proactive strategy centered on early risk stratification, individualized management, and prevention of portal hypertension-related complications.

Despite these advances, important challenges remain before many of these techniques can be routinely incorporated into clinical practice. Although early studies have found encouraging technical success and safety, larger prospective multicenter studies are needed to validate long-term outcomes and establish standardized training pathways. Endo-hepatology represents one of the most significant recent advances in portal hypertension management by integrating portal pressure measurement, liver biopsy, and vascular interventions into a comprehensive model of care. As evidence continues to evolve, endo-hepatology has the potential to redefine the diagnosis and management of portal hypertension and move the field toward more personalized, multidisciplinary, and prevention-focused care.

ACKNOWLEDGEMENTS

The authors thank all collaborators and mentors who provided valuable guidance and feedback during the preparation of this manuscript.