Is Endoscopic Ultrasound-Guided Hepaticogastrostomy Safe and Effective after Failed Endoscopic Retrograde Cholangiopancreatography?—A Systematic Review and Meta-Analysis

Background/Objectives: Endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS) has emerged as an alternative option for biliary drainage in cases of failed endoscopic retrograde cholangiopancreatography (ERCP). Limited data exist on the safety and efficacy of EUS-HGS. In this comprehensive meta-analysis, we aim to study the safety and efficacy of EUS-HGS in cases of failed conventional ERCP. Methods: Embase, PubMed, and Web of Science databases were searched to include all studies that evaluated the efficacy and safety of EUS-HGS. Using the random effect model, the pooled weight-adjusted event rate estimate for clinical outcomes in each group were calculated with 95% confidence intervals (CIs). The primary outcomes were technical and clinical success rates. Secondary outcomes included overall adverse events (AEs), rates of recurrent biliary obstruction (RBO), and rates or re-intervention. Results: Our analysis included 70 studies, with a total of 3527 patients. The pooled technical and clinical success rates for EUS-HGS were 98.1% ([95% CI, 97.5–98.7]; I2 = 40%) and 98.1% ([95% CI, 97.5–98.7]; I2 = 40%), respectively. The pooled incidence rate of AEs with EUS-HGS was 14.9% (95% CI, 12.7–17.1), with bile leakage being the most common (2.4% [95% CI, 1.7–3.2]). The pooled incidence of RBO was 15.8% [95% CI, 12.2–19.4], with a high success rate for re-intervention (97.5% [95% CI, 94.7–100]). Conclusions: Our analysis showed high technical and clinical success rates of EUS-HGS, making it a feasible and effective alternative to ERCP. The ongoing development of dedicated devices and techniques is expected to make EUS-HGS more accessible and safer for patients in need of biliary drainage.


Introduction
Endoscopic retrograde cholangiopancreatography (ERCP) is the gold standard therapy for biliary obstruction for a variety of benign and malignant pancreaticobiliary disorders, with a success rate reaching up to 95% [1][2][3].However, in cases with surgically altered anatomy or malignant duodenal obstruction, it can be very challenging and has a failure rate ranging from 5-7% in achieving biliary drainage [4].For years, the standard alternative in this situation was limited to percutaneous transhepatic biliary drainage (PTBD).However, in recent years, endoscopic ultrasound (EUS)-guided biliary interventions have emerged as effective alternate treatment options.EUS-guided biliary drainage (EUS-BD) techniques include EUS guided rendezvous ERCP, EUS-guided choledochoduodenostomy, and EUS-guided hepaticogastrostomy (EUS-HGS) [5].
Among the EUS-BD techniques, EUS-guided hepaticogastrostomy (EUS-HGS) has gained popularity as a novel drainage technique that provides biliary decompression from the left intrahepatic duct (IHD) to the stomach [6].This method leverages the power of endoscopic ultrasound to access the biliary system, offering distinct advantages over other conventional techniques.
Despite its overall efficacy and safety, clinicians continue to have significant concerns for bile leak and stent migration with EUS-HGSs [7,8].Therefore, EUS-guided antegrade stenting (EUS-AGS) has emerged as a valuable alternative to EUS-HGS, particularly for patients with an inaccessible ampulla, due to its potential to establish normal bile flow [9].Despite the fact that these techniques have been around for almost a decade, there are concerns around the safety and efficacy of these modalities.Therefore, we have conducted a systematic review and meta-analysis to evaluate the efficacy and safety of EUS-HGS and EUS-AGS in cases of unsuccessful conventional ERCP.

Search Strategy and Study Eligibility
Two independent reviewers (S.A. and M.M.) identified studies published before 1 October 2023, that reported on the outcomes of EUS-HGS and EUS-AGS in cases of unsuccessful conventional ERCP.We systematically searched the online MEDLINE, Embase, Cochrane, and Scopus databases using key words in different combinations: (EUS, Endoscopic Ultrasound, Ultrasound) and (Hepaticogastrostomy, biliary drainage, anterograde stenting).Additionally, according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA), we screened the reference lists of the articles and corresponded with study investigators [10].There was no restriction based on language as long as study outcomes were mentioned in the text.A third reviewer (O.T.) resolved any disagreement.

Study Inclusion and Exclusion
The inclusion criteria for studies in this analysis were as follows: (1) Prospective or retrospective studies with a study population comprising patients with biliary obstruction.(2) Studies involving the use of EUS-HGS or EUS-AGS as the primary intervention.
(3) Evaluation of clinical safety and efficacy as the primary outcomes.
Studies were excluded if they were case reports, case series, animal studies, editorial articles, meta-analyses, review articles, or had sample sizes smaller than 10.Studies without relevant clinical data on clinical success or adverse events were also excluded.

Data Extraction and Quality Assessment
All relevant data were extracted according to a table independently predefined by S.A. and M.M.The following parameters were extracted: first author, year of publication, country, study design, patient demographics, stent type, cause of prior failed ERCP, stent patency time, technical success, functional success, and outcomes of interest.Using the Newcastle-Ottawa Scale, the methodological quality of the included cohort studies was assessed independently by two investigators (S.A. and O.T.).In the case of a discrepancy, a third independent individual (A.M) was consulted.

Definitions of Outcomes
The endpoint outcomes include stent patency, stent occlusion, and overall adverse events (AEs).The American Society for Gastrointestinal Endoscopy lexicon was used for the grading of the severity of procedural AEs with endoscopy [11].Technical success of both EUS-HGS and EUS-AGS was generally defined as the successful biliary drainage as planned.Clinical success was defined as a reduction in serum bilirubin level by more than 50% at 2 to 4 weeks.Recurrent biliary obstruction was considered in case of stent migration, occlusion, or malignancy invasion of stent.Our primary outcome is technical and clinical success rate.Secondary outcomes include stent patency, stent occlusion, and adverse events.

Data Synthesis and Statistical Analysis
We used R version 3.2.3(R Project for Statistical Computing) with Meta and Metaprop packages for all analyses.Using the Freeman-Tukey double arcsine transformation (FTT) method, the pooled, weight-adjusted event rate estimate for the clinical outcomes in each group was calculated using the Metaprop package.Continuity correction of 0.5 in studies with zero cell frequencies was used.Between-study heterogeneity was assessed using the Cochrane Q-statistic (I 2 ), which represents the percentage of total between-study variation that cannot be attributed solely to chance.Between-study heterogeneity was rated as low if 25% < I 2 ≤ 50%, moderate if 50% < I 2 ≤ 75%, and high if I 2 > 75%1.A leave-1-out metaanalysis was performed to assess the influence of the outcome by excluding each study and identifying influential studies that may contribute to heterogeneity.A subgroup analysis was performed for studies that reported the outcomes of EUS-HGS with anterograde stent.Statistical tests were 2-sided and used a significance threshold of p < 0.05.The assessment of publication bias was investigated by evaluation of funnel plot asymmetry and sensitivity analysis.In addition to the ethical standards of the competent institution for human subjects, this meta-analysis was conducted in compliance with the Helsinki Declaration [12].

Literature Search and Study Characteristics
A total of 3276 unique records were identified according to the above search strategy.After title and abstract screening, 70 studies with a total of 3527 patients were included in the study.PRISMA flowchart illustrates our selection process as shown in Figure 1.Supplementary Table S1 shows the baseline characteristics of the included studies and their quality analysis.Of these studies, 53 were from Asia.The study design was prospective in 19 studies, and 23 were multi-center studies.Among the included studies, 43 were of good quality, 21 were of fair quality, and 6 were of poor quality.Supplementary Table S2.

Baseline Characteristics of Patients and Qualitative Procedure Outcomes
Table 1 shows the baseline characteristic of patients in the studies included and procedure outcomes including: gender, age, underlying cause of obstruction, reason for prior unsuccessful ERCP, overall survival, stent patency time, median procedural time, type of stent, and location of stricture.While 61 studies included only patients with malignant obstruction, 8 included mixed malignant and benign obstruction, and 1 study included only benign obstruction.The most common cause of obstruction was pancreatic cancer.Distal bile strictures were the most common location of stricture.Metal stents were the most commonly used type.

Assessment of Publication Bias and Sensitivity Analysis
A funnel plot of included studies is shown in Figure 4, which indicates no publication bias.The symmetry of the plot around the central line suggests an even distribution of study results, implying that both positive and negative outcomes were equally likely to be published.This balanced spread of effect sizes across studies of varying sizes supports the conclusion that there is no selective publication bias affecting the meta-analysis results.Additionally, the influence of a single study on the overall meta-analysis estimate was investigated by omitting one study at a time.The omission of any study resulted in no significant difference, indicating that our results were statistically reliable.

Discussion
In this comprehensive systematic review and metanalysis of 70 studies with 3643 patients, the pooled clinical success of EUS-HGS after unsuccessful ERCP was 90.9% and the technical success rate was 98.1%.Our analysis showed that the pooled incidence of AEs with EUS-HGS was 14.9% with bile leak being the most common AE at 2.4%.To our knowledge, this is the largest analysis including 70 studies focused on EUS-HGS.Our analysis provides valuable insights into the efficacy and safety of EUS-HGS with or without EUS-AGS when conventional ERCP fails or is not feasible.
Biliary obstruction is a challenging medical condition that can often necessitate a wide variety of interventions via endoscopic retrograde cholangiopancreatography (ERCP).When ERCP is unsuccessful, endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS) emerges as a promising alternative for biliary drainage [82].Numerous studies have provided insights into the efficacy and safety of EUS-HGS as an option for biliary drainage [83][84][85].The high success rates make EUS-HGS a viable alternative when conventional ERCP is not feasible, particularly in cases involving surgically altered anatomy or inaccessible papilla.Our meta-analysis showed results further supporting this expanding body of evidence, with a pooled clinical success rate of 90.9%, and a pooled technical success rate of 98.1%.Compared to endoscopic retrograde biliary drainage (ERBD) and percutaneous transhepatic biliary drainage (PTBD), EUS-HGS shows comparable success rates [86,87].Moreover, EUS-HGS has the added advantage of being accessible in cases where ERBD is not feasible due to anatomical challenges.
Stent patency in EUS-HGS is a crucial aspect of its long-term effectiveness.Although theoretically, stent patency might be longer in EUS-HGS than in ERBD, various factors influence the duration of patency [88].Reported stent patency durations for EUS-HGS have varied widely, ranging from 62 to 402 days.While EUS-HGS may have fewer instances of tumor ingrowth or overgrowth, it can be more susceptible to stent migration and clogging, potentially shortening stent patency.In our review which included 70 studies, there was a significant variation in the stent patency duration, ranging from 31 to 771 days with a pooled average stent patency of 155 days.The location and degree of biliary stricture, presence of gastric or duodenal obstruction, type and length of the stent used, presence of liver metastasis, and other factors all contribute to the stent patency of EUS-HGS.
In the studies included in our meta-analysis, we observed a notable variation in the reported types of stents utilized.Out of the 70 total studies, 36 (51.4%) mentioned the deployment of metal stents, while 8 studies (11.4%) specifically indicated the use of plastic stents.Additionally, 24 studies (34.3%) scrutinized the utilization of both metal and plastic stents.In contrast, 2 studies (2.9%) did not provide information regarding the type of stent used.
A recent prospective study compared stent patency between EUS-guided biliary drainage (EUS-HGS and EUS-choledochoduodenostomy) and ERBD, showing that EUSguided drainage had significantly longer stent patency [6].However, it is essential to consider patient survival when interpreting these results, as many patients with biliary obstruction have a short survival time.Shorter survival may reduce the likelihood of observing stent dysfunction because patients may not live long enough for the stent to fail.This distinction is crucial for understanding the actual efficacy and reliability of the stent.
EUS-HGS has been shown to produce fewer procedure-related adverse events than PTBD, making it a safer alternative [6].The overall previously reported rate of adverse events in EUS-HGS is approximately 18%, with common adverse events including abdominal pain, self-limiting pneumoperitoneum, bile leak, cholangitis, and bleeding.However, in rare cases, serious adverse events like perforation, intraperitoneal stent migration, and mediastinitis have been reported.Notably, our analysis indicates a pooled adverse events incidence rate of 14.89%, with the most frequently encountered adverse events being bile leakage (2.4%), bleeding (1.3%), and peritonitis (1.27%).
Compared to EUS-HGS, EUS-choledochoduodenostomy (EUS-CDS) was shown to result in less early adverse events and shorter procedure time [20].This suggests that EUS-CDS might be a safer option for novice practitioners, while EUS-HGS should be reserved for experienced operators [89].The learning curve for EUS-HGS is steep, and it is a technically challenging procedure.Studies suggest that achieving proficiency in EUS-HGS may require a significant number of cases, with some reports indicating that more than 33 cases may be needed to reach a plateau in the learning curve [52].While EUS-HGS is still technically challenging, one approach to mitigate the learning curve is the conversion of PTBD to EUS-HGS for beginners [90].This transition can offer several advantages, including the ability to identify the optimal puncture site in the intrahepatic duct via opacification through a PTBD catheter.Furthermore, it allows practitioners to become more familiar with EUS-HGS while potentially reducing the risk of adverse events, such as cholangitis or bile leak.
Despite its advantages, EUS-HGS has several limitations and challenges that need to be considered [6].Some of these limitations include the technical complexity of the procedure, the lack of dedicated devices for EUS-HGS, the risk of serious adverse events, and the need for skilled practitioners.Additionally, there are technical challenges in draining the right liver in cases of bilateral stenosis and difficulties in patients with a non-dilated biliary system [91].
It is imperative to address the limitations inherent in our meta-analysis.A substantial portion of the studies incorporated in our analysis adopted a retrospective design.This approach has the potential to result in an overestimation of technical and clinical success rates, particularly when populations of patients initially assigned to EUS-HGS or EUS-AGS were subsequently transitioned to an alternative method, such as EUS-choledochoduodenostomy, and labeled as such.Furthermore, several essential parameters that might be of interest, such as the stratification of adverse events and outcomes based on common bile duct size or stent dimensions, were frequently omitted in the studies we reviewed.In recent years, various new devices, including dedicated stent systems, have been developed to make EUS-HGS more accessible [36].These innovations aim to simplify the procedure, increase its success rate, and decrease the risk of adverse events.Continued development in this area is essential to improve the safety and feasibility of EUS-HGS.

Conclusions
EUS-HGS has emerged as a valuable alternative for biliary drainage when conventional ERCP is not feasible.The technique has shown high technical and clinical success rates and potentially longer stent patency compared to ERBD [6].However, it is not without its challenges, including a steep learning curve, the need for skilled practitioners, and potential risks of adverse events.The ongoing development of dedicated devices and techniques is expected to address these challenges, making EUS-HGS more accessible and safer for patients in need of biliary drainage.

Table 1 .
Comprehensive characteristics of included studies.

Table 2 .
Details of clinical and technical success rates and adverse events associated with EUS-HGS and HGAS.