Endo‐hepatology: The changing paradigm of endoscopic ultrasound in cirrhosis

Short abstract

Content available: Author Interview and Audio Recording

Watch the interview with the author.

Listen to an audio presentation of this article.

INTRODUCTION

Cirrhosis and portal hypertension encompass a broad spectrum of diseases associated with a myriad of manifestations and complications. The present clinical management of patients with advanced liver disease and portal hypertension, including diagnosis, surveillance, and management, relies on a multidisciplinary approach, including specialists in hepatology, interventional radiology, surgery, pathology, and interventional endoscopy. Recent innovations in EUS have expanded the role of endoscopy in diagnosis and management of patients with liver disease. Endoscopic hepatology, or “endo‐hepatology,” is an emerging field comprising innovative EUS‐guided modalities aimed at providing diagnostic and/or therapeutic interventions for patients with cirrhosis and other diseases of the liver. ¹ For example, a patient undergoing an upper endoscopy for variceal screening can be simultaneously offered EUS‐guided LB and EUS‐guided portal pressure measurements. This concept of a “one‐stop‐shop” can potentially simplify health care delivery, improve accessibility, and enhance patient satisfaction and follow‐up.

Endo‐hepatology plays an important role in the clinical management of patients with cirrhosis, those with pre‐cirrhotic diseases of the liver, and patients with portal hypertension. The present realm of EUS‐guided interventions in liver disease encompasses the diagnosis of liver pathology and fibrosis/cirrhosis via liver biopsy (LB), measurement of portal pressure gradient (PPG), and management of gastric varices (GVs). ² Understanding the role of these innovative modalities, as well as the limitations in the current literature, will facilitate the appropriate adoption of these practices and stimulate further research in the field. In this article, we elucidate the present knowledge on the role of EUS in the diagnosis and management of patients with advanced liver disease and portal hypertension.

DIAGNOSTIC UTILITY OF EUS

EUS‐guided LB

Histopathological examination remains the gold standard for the diagnosis of liver diseases and cirrhosis despite the various advancements in the noninvasive assessment of fibrosis. Conventional methods for LB include percutaneous LB (PLB) and transjugular LB (TJLB). For TJLB, the right internal jugular vein is accessed, and the right hepatic vein is most frequently cannulated. This approach, although safe, can be associated with risks for arrhythmias, especially ventricular tachycardias, as the guidewire crosses the right atrium during the procedure. Further, TJLB can be complex or prohibited in patients with jugular vein thrombosis, hepatic vein obstruction, or those with hydatid cysts. ³

PLB is an effective, low‐cost method performed using palpation/percussion‐guided or ultrasound (US)‐guided techniques. It is, however, associated with potential adverse events, ranging from pain in mild cases to death in the setting of procedural complications. ⁴ The efficacy and safety of this technique relies on several factors in combination, including patient cooperation, use of US, type of needle used, and the experience of the operator. The use of US guidance reduces the risk of bleeding and major complications and is therefore recommended in all PLB procedures. ⁵ PLB is contraindicated in uncooperative patients, those with severe coagulopathy or extrahepatic biliary obstruction, and patients with ascites or morbid obesity.

Considering these traditional approaches for LB, transgastric or transduodenal EUS‐guided LB offers some distinct advantages. The endoscopic approach provides access to both lobes of the liver, facilitating biopsy of even small lesions. ⁶ This may allow for improved target‐directed sampling of the liver lesions compared with TJLB or PLB. Moreover, it can be accomplished simultaneously with endoscopy for variceal screening, thereby reducing the need for additional interventions in this complex patient population. In addition, there is no requirement for intravascular access or catheter placement (Figure 1). However, these advantages should be balanced with the need for deeper sedation, higher cost, and presently, the limited availability of this procedure to centers of expertise.

FIGURE 1.

Summary of the present applications of EUS‐guided therapies in patients with cirrhosis. Blue: strengths of the technique; red: present limitations.

The present data regarding the comparative efficacy of EUS‐guided LB with PLB and/or TJLB are limited. A systematic review of 656 patients from four retrospective studies and one prospective analysis compared the three modalities for the adequacy of outcomes as determined by the number of complete portal triads (CPTs), total specimen length, and length of the longest piece. ⁶ The adequacy rate for EUS‐guided biopsy was 93.5% compared with 97.6% with TJLB and 98.3% for PLB. All the included studies had a low risk of bias. Similarly, in a second systematic review, the pooled successful histological diagnosis rate for specimens obtained via endoscopic guidance was 93.9% associated with an overall adverse event rate of 2.3%. However, there was significant heterogeneity (75.3%) in the included studies. ⁷ A recent randomized trial found the total specimen length of LB was significantly shorter for patients with EUS‐guided LB compared with PLB. ⁸ Although PLB was associated with higher pain during the first hour postprocedure, it was associated with lower cost. That said, several questions were raised concerning the variability in techniques reported (and lack of standardization thereof) for the acquisition of tissue with EUS‐guided LB. ⁹ Interestingly, recent preliminary results of a multicenter randomized trial from India performed in 48 patients found EUS‐guided LB had significantly better aggregate specimen lengths and more CPTs compared with PLB. ¹⁰

There are several areas undergoing active exploration for the optimization of EUS‐guided LB, including needle gauge, unilateral or bilateral lobe biopsies, number of passes, need for suction, and others. ¹¹ The current literature supports the feasibility of endo‐hepatology as a diagnostic modality with an acceptable safety profile and high technical success rate. However, it is important to note that the literature is driven by experience from tertiary care hospitals and as a result, community‐based studies are required to establish its broad utility and reproducibility. ¹²

EUS‐guided portal pressure measurement

Portal hypertension, defined as an elevated hepatic venous pressure gradient (HVPG) > 5 mm Hg, drives much of the disease‐related morbidity and mortality in patients with cirrhosis and advanced liver disease. ¹³ The accurate assessment of HVPG can facilitate individualized screening and surveillance protocols, guide decision‐making, and support disease prognostication. The present methods for assessing HVPG rely on measuring the free hepatic vein pressure (FHVP) and the wedged hepatic vein pressure (WHVP) through transjugular hepatic vein cannulation. The HVPG is the difference between the FHVP and WHVP. The WHVP has been shown to be a reliable surrogate marker for portal venous pressures (Figure 2A). ¹⁴ In terms of procedure‐specific limitations, the transjugular hepatic vein cannulation method requires exposure to radiation via fluoroscopy and may not be suitable for the evaluation of patients with presinusoidal or prehepatic portal hypertension.

FIGURE 2.

Hepatic venous pressure measurement using (A) transjugular method and (B) EUS method.

The EUS‐guided PPG (EUS‐PPG) measurement is emerging as a reliable method to assess HVPG. In this method, via transgastric and/or transduodenal routes, pressure measurements are obtained from the hepatic vein (HVP) and from the intrahepatic portal vein (PVP). The difference between these measurements provides the PPG (PVP – HVP = PPG) (Figure 2B). This endoscopic method does not rely on any surrogate pressures for HVPG measurement. While the first human study was done in 2017, there have been four studies in total that have assessed the feasibility of EUS for portal pressure measurement (Table 1). ⁹ , ¹⁵ , ¹⁶ , ¹⁷ Advantages of EUS‐PPG include the direct assessment of the portal venous pressures, the absence of exposure to ionizing radiation, and its role in evaluating patients with presinusoidal disease. These endoscopic techniques have the potential to facilitate the assessment of PPG during screening and surveillance endoscopies, thereby providing real‐time assessment of luminal variceal status, as well as the prognostication of variceal risk in patients with clinically significant portal hypertension.

TABLE 1.

The various studies evaluating the role of EUS‐guided portal pressure measurement.

Author, country, year	Type of study	Primary outcome	No. of patients	Findings	Study strengths	Study limitations
Huang, United States, 2017 16	Prospective cohort	Assess feasibility of EUS‐PPG and correlation with clinical features	28	Successful procedure: 100%	First human pilot study demonstrating PPG measurements and correlation with clinical features	No comparator arm in the study
				Adverse events: none reported
				Primary outcome: PPG: 1.5‐19 mm Hg and correlated with presence of varices (p = 0.002), thrombocytopenia (p = 0.036), and portal hypertensive gastropathy (p = 0.002)
Zhang, China, 2020 17	Prospective cohort	Comparison of the accuracy of EUS‐PPG with transjugular HVPG	12	Successful procedure: 11 (91.7%), transjugular successful in 9 (75%) a	Only study that compares EUS and transjugular methods	Small sample size
				Primary outcome: mean EUS‐PPG: 18.07 ± 4.32; transjugular HVPG: 18.8 ± 3.43 (p < 0.01)
Choi, United States, 2022 15	Retrospective cohort	Comparison of EUS‐PPG measurements with histological fibrosis stage and various surrogate clinical markers for severity of CLD	64	Successful procedure: 100%	Largest study to date	Single‐center, retrospective study
				Adverse event: no serious events
				Primary outcome: in 29 patients with EUS‐PPG > 5 mm Hg, significant correlations with clinical cirrhosis (p < 0.001) and clinical portal hypertension (p < 0.001)
Hajifathalian, United States, 2022 9	Prospective cohort	Assess the safety, feasibility, and technical success of simultaneous EUS‐PPG and EUS‐guided LB	24	Successful procedure: EUS‐PPG accomplished in 23 patients (95.8%)
				Adverse outcome: abdominal pain in one patient
				Primary outcome: association between EUS‐PPG and transient elastography (p = 0.011) and fibrosis‐4 score (p = 0.559); no significant correlation between fibrosis stage on histology and the measured EUS‐PPG

^a HVPG was not feasible in two patients with Budd–Chiari syndrome and one patient with hepatic vein shunts.

Moreover, in patients with certain etiologies of chronic liver disease, EUS‐PPG may yield a more accurate estimation of the degree of portal hypertension than via transjugular hepatic vein cannulation. For example, in a study of patients with decompensated non‐alcohol‐associated steatohepatitis‐related cirrhosis, the WHVP was shown to not correlate well with the directly assessed PVP. This finding has been attributed to a presinusoidal component of the disease, generated by periportal fibrosis and ongoing portal venous endothelial damage. ¹⁸ In contrast, the generalizability of most studies is limited by small sample sizes and the requirement for expertise in EUS, which is at present available only in certain tertiary care hospitals in the United States. Further, it is important to note that sedation has been shown to impact the interpretation of HVPG, as confirmed by prior studies. ¹⁹ , ²⁰ EUS‐PPG requires deep sedation. To date, there have been no studies that have assessed the impact of sedation on EUS‐PPG. Although this currently stands as a limitation, future studies on EUS‐PPG should highlight the role of sedation and its possible impact on PPG interpretation. Another important limitation is the lack of reference standards for EUS‐PPG interpretation at present. Future studies on EUS‐PPG should evaluate its outcomes in comparison with the transjugular route for measuring HVPG and study its role in assessing disease progression and prognostication, particularly given the widespread availability of well‐established noninvasive modalities for the assessment of liver stiffness and fibrosis.

THERAPEUTIC UTILITY OF EUS

EUS‐guided management of GVs

Although less common than esophageal varices, GVs are reported in 10% to 20% of patients with cirrhosis. GV bleeding is associated with a high risk of rebleeding, and few highly efficacious treatment strategies exist. Although the Sarin classification has been the most widely adopted GV classification system, numerous other classification systems have emerged. Based on a recent practice update, the Saad Caldwell definition now classifies GVs as (1) lesser curvature GVs if located on the lesser curvature of the stomach, (2) cardiofundal GVs if located on the posterior or the greater curvature of the cardia, or (3) distal GVs if located in the antrum or the gastric body. ²¹

The established methods for management of GVs include medical (pharmacological) management, endoscopic therapies, and IR‐guided procedures, such as transjugular intrahepatic portosystemic shunt (TIPS) and balloon retrograde transvenous obliteration (BRTO). Traditional endoscopic therapies include intravariceal injection of thrombosis‐inducing materials such as acrylate glue or thrombin. These endoscopic therapies are technically feasible and successful in hemostasis but are associated with significant rates of rebleeding and have a rare risk of embolization (<1%). ²² Although IR‐guided procedures have shown promising results in managing GVs, they remain associated with certain limitations. TIPS can be complicated by encephalopathy or shunt dysfunction. The risk of encephalopathy is lower in BRTO, but it requires the presence of a gastrorenal shunt and does not divert the blood flow. Further, BRTO can actually increase the portal pressures, leading to worsening of esophageal varices or ascites. ²³ , ²⁴

EUS with Doppler has emerged as a promising modality for GVs, providing real‐time evaluation and allowing for more targeted interventions (Figure 3). EUS‐guided glue injections may even require a lower quantity of the glue and are associated with less frequent rebleeding episodes than traditional endoscopic therapies. ²⁵ Another emerging technique is EUS‐guided coil embolization of GVs, which entails the injection of coils into variceal nests to promote thrombosis. This method has demonstrated high rates of clinical and technical success with a favorable safety profile. ²⁶ Coil embolization can be performed alone or coupled with glue injection. It is hypothesized that packing a variceal nest with coils provides a scaffold for the injectate, thereby potentially reducing the risk of systemic embolization. A recent systematic review of 851 patients from 23 studies compared the outcomes of EUS‐guided GV interventions and endoscopic glue injections. ⁷ Obliteration of the GVs was improved using EUS‐guided therapies compared with the standard endoscopic intravariceal glue injection. In another systematic review of 573 patients from 11 studies, combined EUS‐guided coiling and glue injection was found to be superior to either of the procedures alone. ²⁷ Moreover, combined therapy had a lower incidence of adverse events when compared with the individual procedures.

FIGURE 3.

EUS‐guided management of GVs. (A) Large fundal GVs seen on retroflexion. (B) EUS‐guided varicealography showing significant Doppler flow. (C) EUS‐guided coil embolization of GVs. (D) Significant reduction in Doppler flow after EUS‐guided coil embolization.

In keeping with the limitations of EUS‐guided LB, the studies of EUS‐guided GV interventions are limited by small sample sizes and the restriction of practice to tertiary academic centers with specific expertise in EUS. Further, limited data exist to guide the optimal number of coils and volume of glue required to achieve variceal thrombosis; these are important subjects of active investigation in the field.

FUTURE DIRECTIONS

Newer emerging applications for EUS‐guided therapies in patients with liver disease include EUS spectrometry, elastography, EUS‐guided portal vein sampling, and endoscopic‐based therapies for hepatic neoplasms (Table 2).

TABLE 2.

Future Directions and Evolving Modalities in Endo‐hepatology.

Future Modalities	Potential Clinical Applications
Diagnostic developments
EUS elastography	Using the proximity of the stomach to the liver, endoscopic elastography can provide higher accuracy compared with the traditional transabdominal probe. This could be particularly beneficial in patients with increased central adiposity or large‐volume ascites where transabdominal elastography may be limited. 28
Contrast‐enhanced EUS	This modality uses US contrast agents that allow for an enhanced depiction of hepatic vascular architecture. Preliminary results show that contrast‐enhanced EUS may improve the detection of left hepatic lobe metastasis from pancreatic neoplasms, when compared with abdominal US and computed tomography of the abdomen. 29 This modality is not currently available in the United States.
Portal venous sampling	Portal vein samples have a higher number of circulating tumor cells from the GI tract, particularly from the pancreas. EUS‐guided interventions can provide access to the portal vein and may aid the study of the pathogenesis, response, and progression of pancreaticobiliary cancers. 30 This modality may also become a prognostic tool to stratify the risk of cancer recurrence or metastasis in GI cancers.
Artificial intelligence	Applications of artificial intelligence for image recognition, clinical predictive modeling, and clinical decision making may be applied to EUS image analysis to highlight potentially suspicious hepatic lesions in patients with cirrhosis, to score surrogates for portal hypertension, and to stratify the risk of bleeding in esophageal and GVs. 31
Therapeutic developments
Lesion‐targeted chemoembolization	EUS‐guided ablation therapies include experimental treatments for liver tumors refractory to percutaneous approaches. These techniques include EUS‐guided needle injection with ethanol and thermal therapy with RFA. An interesting yet unproven therapy proposed is the injection of chemotherapy directly into the portal vein for diffuse diseases of the liver. 32 Similar to IR‐guided TACE, animal models studying the delivery of chemotherapy‐eluting beads in the portal vein are ongoing. 2
EUS‐guided portosystemic shunt	This method has not yet been studied in clinical phases, but its utility has been explored in a porcine modeling study. 33 Improvement in the type/shape of stents, stent delivery modalities, and techniques to mitigate against stent thrombosis is needed.
Gene therapy	A preclinical study of gene therapy using DNA plasmids delivered to the liver via endoscopic retrograde cholangiopancreatography‐guided hydrodynamic injection into the bile duct in live pigs has been reported. 34 Genomic integration and protein expression were observed in targeted liver tissue in animal models surviving to 60 days.

CONCLUSION

Endo‐hepatology is an emerging field encompassing EUS‐guided interventions aimed at facilitating minimally invasive diagnosis and management of patients with liver disease and portal hypertension. As experience and training with these modalities mature, so does the potential for widespread application of endo‐hepatology approaches. Currently, the body of evidence supporting EUS‐guided interventions is driven by tertiary care academic centers. Large‐scale multicenter and community‐based studies are needed to establish the role of endo‐hepatology in everyday practice.

CONFLICT OF INTEREST

Nothing to report.

Singh AD, Bazarbashi AN, Lindenmeyer CC. Endo‐hepatology: The changing paradigm of endoscopic ultrasound in cirrhosis. Clin Liver Dis. 2022;20:209–215. 10.1002/cld.1263