Endohepatology: Arrival at the frontier of interventional endosonography

Abstract

Endohepatology describes the emerging field where diagnostic and therapeutic endoscopic ultrasound (EUS) are used for the diagnosis and management of liver disease and its sequelae. In this editorial we comment on the article by Gadour et al. The spectrum of EUS-guided procedures includes liver parenchymal and lesional biopsy, abscess drainage, treatment of focal liver lesions, diagnosis of portal hypertension and management of gastric varices. The data suggest that the application of EUS to hepatology is technically feasible and safe, heralding the arrival at a new frontier for EUS. More data, specifically randomised trials comparing EUS to interventional radiology techniques, and continued partnership between endoscopy and hepatology are required to see this field establish itself outside expert tertiary centres.

Key Words: Cancer; Cirrhosis; Endoscopy; Endoscopic ultrasound; Liver disease

Core Tip: Endohepatology, where endoscopic ultrasound (EUS) and hepatology are combined, allows for novel and minimally invasive ways to investigate and treat liver disease. These procedures, especially EUS-guided liver biopsy, portal pressure gradient measurement and obliteration of gastric varices, have now been demonstrated to be technically feasible and safe; thus, highlighting the continued expansion, and clinical implications of diagnostic and therapeutic EUS.

INTRODUCTION

For evaluation and management of liver disease and portal hypertension, hepatologists have predominantly turned to interventional radiology (IR) for diagnostic procedures such as liver biopsy and transjugular hepatic venous pressure gradient measurement (HVPG), endovascular treatments for complications of portal hypertension such as transjugular intrahepatic portosystemic shunts (TIPS), and image-guided anti-tumour interventions for hepatocellular carcinoma[1,2]. Endoscopy does have an established role in the management of cirrhosis, but this has historically been confined to the screening, prophylaxis and management of varices[3,4].

Endoscopic ultrasound (EUS) is an established modality for diagnosis and tissue acquisition, but in the last two decades there have been sizable advances in therapeutic EUS, especially as an alternative to surgical intervention on a range of pancreaticobiliary disorders[5,6]. The integration of EUS with hepatology, a field termed “Endohepatology” by Chang et al[7] over a decade ago, has also seen significant expansion. It is attractive to apply EUS to the diagnosis and management of a patient with liver disease, due to the capability of real-time imaging and instrument guidance, and the proximity of the probe to relevant parenchyma and vasculature[8]. The spectrum of EUS applications to hepatology includes diagnostic procedures such as EUS-guided assessment and biopsy of the parenchyma or focal liver lesions, portal hypertension measurement, as well as therapeutic procedures such as abscess drainage, management of gastric varices and intratumoral therapy[9,10].

In this issue of the World Journal of Gastroenterology, Gadour et al[11] undertook a systematic review and meta-analysis of studies evaluating the role of diagnostic and therapeutic EUS in liver diseases. 45 studies were included, 28 of which evaluated the diagnostic role of EUS in parenchymal liver disease, focal liver lesions and portal hypertension, and 17 of which evaluated the therapeutic role of EUS in liver abscess drainage, treatment of liver lesions and treatment of gastric varices.

APPLICATIONS OF EUS IN HEPATOLOGY

For the diagnosis of parenchymal liver disease with EUS-guided liver biopsy (EUS-LB), the overall diagnostic accuracy was reported as 96.6%, and complication rates of 6.2% and 9.6% for use of EUS fine-needle aspirate (FNA) and EUS fine-needle biopsy (FNB). EUS-LB has several perceived advantages compared to the standard percutaneous or transjugular approaches, including reduced patient discomfort due to associated sedation, the ability to perform high-resolution real-time imaging and ancillary diagnostic modalities such as elastography, the ability to sample both lobes of the liver reducing sampling error, and allowing access to liver tissue even in patients with morbid obesity. Various techniques have been reported, but currently the favoured technique is usage of a 19-gauge FNB needle (which has been shown to yield longer specimens and greater number of complete portal triads relative to EUS-FNA), wet suction and three to four actuations with a 3 cm course of needle travel[12]. Adequacy of the histological specimen in terms of specimen length and number of complete portal triads has been reported to be comparable to tissue obtained via percutaneous and transjugular route in a retrospective study[13]. A recent meta-analysis evaluating 33 studies (2098 patients) for EUS-LB reported a pooled diagnostic yield of 95%, and adverse event rate of 3%[14]. It should be noted that EUS-LB carries a higher cost than percutaneous biopsy[15] and remains contraindicated in situations of coagulopathy where transjugular liver biopsy should be performed. Finally, the trend of EUS-LB is against the prevailing paradigm in hepatology of moving towards non-invasive, point-of-care testing, with the recent Baveno VII consensus workshop providing new recommendations that reduce the need for endoscopy and invasive transjugular portal venous pressure measurements[16].

The authors showed that EUS for sampling of focal liver lesions had an overall diagnostic accuracy of 92.4% and a low complication rate of 3.1%, which is in line with a systematic review by Sbeit et al[17] evaluating nine papers (463 patients) reporting a diagnostic yield of 94.8% and a complication rate of 1.5%. The benefits of EUS in this setting as compared to percutaneous approaches include greater detection rate of lesions < 1 cm[18], close approximation of the probe to the target, and the ability to perform real-time imaging guidance of the needle. These benefits have been leveraged for therapeutic purposes in the setting of difficult-to-access lesions to perform EUS-guided liver abscess drainage, for which this review reported an overall technical success rate of 87.5%-100% across four small retrospective studies, as well as EUS-guided ablation/brachytherapy which was described in two case series. However, EUS has limitations, as although segments 1-4 of the liver (caudate and left lobe) are easily accessible, segments 5 and 8 may be difficult and segments 6 and 7 may not be accessible to sample via a transluminal approach. As such, EUS remains a salvage option for targeting focal liver lesions in centres with the available expertise.

Portal hypertension is the driver for the majority of the complications of cirrhosis, and clinically significant portal hypertension [portal pressure gradient (PPG) ≥ 10 mmHg] is associated with the onset of hepatic decompensation and the complications of ascites, variceal haemorrhage or hepatic encephalopathy[3]. The current gold standard for measurement of the PPG is IR-guided transjugular approach to hepatic vein catheterisation and calculating the difference between the wedged hepatic venous pressure and the free hepatic venous pressure; but this has drawbacks including the risk of arrythmia or vascular injury and may be inaccurate in presinusoidal portal hypertension[19]. EUS-guided PPG measurement (EUS-PPG) is an appealing alternative as direct access to the portal vein can be achieved. This technique using a 25 G needle and attached manometer, was first reported in a porcine model by Huang et al[20], with a 100% technical success rate and excellent correlation (R: 0.985-0.99) with the standard IR-guided approach. This meta-analysis reported that in the two human studies involving 40 patients, the EUS-PPG technical success rate was 95.1% with no complications. EUS-PPG has proven technical feasibility and is a seminal procedure for the field of endohepatology, particularly since one can envisage a patient with suspected cirrhosis undergoing EUS-LB, EUS-PPG and variceal screening in a single endoscopic procedure. Indeed a prospective study evaluating same session EUS-PPG and EUS-LB demonstrated feasibility, safety and correlation with clinically evident portal hypertension and non-invasive markers of fibrosis[21]. Barriers to future adoption include the concern that sedation might impact pressure readings, limited availability of expertise, the need for larger scale prospective trials comparing this technique to HVPG, and the need to build awareness of this procedure with the referring hepatologists.

Acute gastric variceal haemorrhage is not as common as esophageal variceal haemorrhage, but has higher mortality and rates of treatment failure[22]. Treatments options are categorised into direct endoscopic treatments (traditionally direct endoscopic glue injection), reduction of portal hypertension (including pharmacologic therapy and TIPS) and obliteration of the variceal network (through balloon-occluded retrograde transvenous obliteration)[23]. EUS-guided interventions include cyanoacrylate glue injection, coil embolization in isolation or combination, and thrombin injection. The pooled analysis of the eleven studies evaluating EUS-guided interventions revealed a technical success rate of 98%, variceal obliteration rate of 94%, a complication rate of 15%, and a rebleeding event rate of 17%. EUS-guided therapy compared to the direct endoscopic glue injection was evaluated in a meta-analysis of 23 studies, which reported superior obliteration rates with EUS-guided therapy, and on subgroup analysis, superiority of a coil-glue combination[24]. Overall, in acute gastric variceal haemorrhage the endoscopic approach is favoured as first-line over endovascular therapies due to the ability to perform hemostatic interventions expediently, and in a fashion that targets areas with bleeding stigmata; moroever EUS-guided intravariceal glue and coil embolization appear to be safer and more efficacious than direct endoscopic injection. The endosonographer must be cognisant of the fact that gastric variceal bleeding is but one manifestation of the larger problem of portal hypertension, and although EUS-guided haemostasis might be achievable for the management of the bleeding episode, the best option for optimal long-term outcomes may require pharmacological (beta-blockade), radiological (TIPS or balloon-occluded retrograde transvenous obliteration) or surgical (liver transplantation) approaches. Furthermore, its role in primary prophylaxis requires further delineation. Sufficient expertise is also required, and with these factors in mind, the European Society of Gastrointestinal Endoscopy suggests that EUS-guided management of bleeding gastric varices combining coil and cyanoacrylate injection may be used in centers with expertise and familiarity with this technique[25].

In this systematic review and meta-analysis, Gadour et al[11] are to be commended in reviewing articles covering a wide spectrum of EUS applications in hepatology, but there are some limitations. The studies are heterogeneous and limited in sample size. Most studies were conducted in single expert centres limiting generalisability, and most studies lacked a control arm, raising a concern for bias. The authors also did not include papers reviewing other EUS procedures relevant to liver disease, including EUS-guided elastography, which adds minimal time to a standard EUS examination, is advantageous in patients with ascites and obesity, and has similar accuracy to percutaneous elastography[26].

CONCLUSION

This article adds to the growing body of evidence that EUS in hepatology, or endohepatology, has a wide spectrum of applications, a high technical success rate, and a low complication rate. The idea of an endoscopist being able to carry out diagnosis and prognostication of liver disease with EUS without a need to outsource to radiology, first conceptualised over ten years ago, has now been proven to be feasible and safe.

How do we achieve greater adoption of these techniques? Technical feasibility and safety are not the only criteria for adoption; larger scale randomised controlled trials are required to demonstrate superiority, or at least non-inferiority with the standard-of-care IR alternatives. There is also the question of who should perform these procedures. This is a nascent field and there are no defined standards for attainment and maintenance of competency[27]. Whilst it may be more feasible for advanced endoscopists to add these endohepatology procedures to their skillset and practice workflow, the complex nature of liver disease and need for a longitudinal, holistic approach means that it is critical that the treating hepatologist is at the helm in the management of these patients and that it is acknowledged that an endoscopic approach, whilst technically feasible, may not be the most appropriate approach for the patient. Finally, the data for these procedures is from tertiary academic centres with specific expertise in complex EUS, and therefore implementation in a real-world and/or community setting may be limited.

This is an exciting time for EUS, and the field of endohepatology has truly marked the arrival at the frontier of interventional endosonography. These novel procedures have the capability to improve the journey and outcomes of a patient with liver disease. Further data and continued partnership of hepatologists and endosonographers are required for us to see adoption beyond selected patient groups in expert centres.