Use of endoscopic ultrasound-guided biliary drainage as a rescue of re-intervention after the placement of multiple metallic stents for malignant hilar biliary obstruction

概要

Introduction
Malignant hilar biliary obstruction (MHBO) is caused by direct invasion of various malignant tumors or by liver and lymphatic metastases. Given the anatomical complexity of MHBO, endoscopic stenting for MHBO requires a high level of skill and has a high frequency of stent failure and cholangitis.(Uppal and Wang, 2015, Lee et al., 2014) Self-expandable metallic stents (SEMSs), which have a longer patency period than plastic stents (PSs), are commonly used for biliary drainage in MHBO. (Xia et al., 2020, Dumonceau et al., 2018) Indeed, drainage of both liver lobes, rather than one, has been demonstrated to result in longer stent patency, (Xia et al., 2020, Ashat et al., 2019) and drainage of >50% of the liver volume is associated with longer patient life expectancy.(Vienne et al., 2010) Therefore, recently, there has been a trend toward multi-stenting with SEMS to drain both liver lobes.

The main re-intervention (RI) approach for RBO after placing uncovered SEMS is the addition of a new stent. Endoscopic transpapillary re-intervention (ETP-RI) (Okuno et al., 2019, Son et al., 2018, Hong et al., 2017, Inoue et al., 2016, Fujii et al., 2013, Ridtitid et al., 2010, Law and Baron, 2013) is a standard method in this situation, but its technical and clinical success rates are 76.3%–100% and 51.7%–100%, respectively, (Okuno et al., 2019, Son et al., 2018, Hong et al., 2017, Inoue et al., 2016, Fujii et al., 2013, Ridtitid et al., 2010) which are all results from retrospective studies, and these methods have not been established yet. Notably, endoscopic ultrasound-guided biliary drainage (EUS-BD) is a widely used biliary drainage method that does not cross the ampulla of Vater or bile duct jejunal anastomosis. Recently, RI with EUS-BD (EUS-RI) for RBO after placing SEMS for MHBO has been investigated in several retrospective studies; (Nakai et al., 2019, Minaga et al., 2017, Ogura et al., 2017, Kanno et al., 2017, Park et al., 2010, Park et al., 2013) however, the sample size was small (5–30 cases) in these studies, and only short-term results were reported. To date, the long-term results (including stent patency) and procedural and clinical success rates of EUS-RI remain unclear. This retrospective study aimed to identify a suitable method of RI, evaluate the utility of EUS-RI, and compare EUS-RI and ETP-RI for RBO after multiple SEMS placement.

Methods
Patients
This study retrospectively reviewed patients with stent failure requiring RI after multi-stenting with uncovered SEMS for MHBO at our institution from October 2017 to April 2021. The eligibility criteria were as follows: 1) unresectable MHBO, 2) >2 uncovered SEMS inserted for MHBO through endoscopic or percutaneous routes, 3) elevated hepatobiliary enzyme and bilirubin levels resulting in stent failure (RBO), and 4) RI was necessary to treat RBO.

If RBO occurred, ETP-RI was the first choice in all cases. When ETP-RI was difficult or when it was clinically impossible to drain with ETP-RI alone, EUS-RI was performed. Percutaneous transhepatic biliary drainage (PTBD) was performed when draining was difficult with both ETP-RI and EUS-RI. If some bile ducts could not be technically drained despite stenting with ETP-RI, they were treated as technical failure cases of ETP-RI, and EUS-RI was added. This study was conducted per the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan and approved by the Institutional Review Board (IRB) of the National Cancer Center Hospital in Japan (IRB number; 2018-149).

Results
During this study period, multi-stenting for MHBO was performed in 180 patients, of whom 49 (27.2%) received ETP-RI. Baseline diseases were cholangiocarcinoma (51.0%, 25/49), gallbladder carcinoma (22.5%, 11/49), pancreatic cancer (14.3%, 7/49), and others (12.2%, 6/49). Bismuth types II, III, and IV were observed in 36.7% (18/49), 53.1% (26/49), and 10.2% (5/49) of the patients, respectively.

The technical success rate of ETP-RI was 69.4% (34/49). Among the 49 patients who underwent ETP-RI, the procedure was unsuccessful in 15 patients. All these cases were converted to EUS-RI. The target bile duct drainage was difficult in 80.0% (12/15) of the cases, and the transpapillary approach was impossible owing to duodenal stenosis in 20.0% (3/15). Fifteen patients underwent EUS-RI, and the technical success rate of EUS-RI was 86.7% (13/15).

The TRBO after RI was 84 days (IQR: 46.0–113.0, 95% CI: 52.0–116.0) for ETP-RI and 212 days (IQR: 186.0–246.0, 95% CI: 157.3–266.7) for EUS-RI, indicating that the EUS-RI group had a significantly longer TRBO than the ETP-RI group (P=0.01) (Fig. 3). In univariate analysis, only EUS-RI showed a significant difference (hazard ratio [HR], 5.46; 95% CI: 1.23–24.24; P=0.03).
Multivariate analysis for TRBO after RI was performed with the initial stenting route and RI method as factors. As a result, EUS-RI was the only independent factor contributing to prolonged TRBO (HR, 4.48; 95% CI: 1.01–19.91; P=0.04) (Table 2).

Discussion
In this study, EUS-RI was safely performed with a high technical success rate as a rescue after technical failure of ETP-RI following multi-stenting with SEMS for MHBO. Furthermore, TRBO after RI for multi-stenting with SEMS for MHBO was significantly longer for EUS-RI than for ETP- RI (84 vs. 212 days, P=0.01), and EUS-RI was the only independent factor associated with TRBO after RI (HR=4.48).

Only one retrospective study mentioned EUS-RI after multiple SEMS placement for MHBO(Kanno et al., 2017). In other reports on EUS-RI, the exact number of multi-stenting cases could not be ascertained. In this study, the technical and clinical success rates of EUS-RI were 86.7% and 100%, respectively, which are comparable to those previously reported.

EUS-RI could be superior to ETP-RI because it does not cross the tumor and is less likely to cause RBO owing to subsequent tumor exacerbation. In this study, 89.8% (44/49) of cases involved tumor ingrowth or overgrowth; thus, the significantly longer TRBO of EUS-RI might have occurred because the drainage was performed distal to the tumor. Additionally, as an RI technique, EUS-RI is superior in terms of procedural feasibility, as it allows direct drainage of the target bile ducts without having to pass through the mesh of the existing stent. Notably, even when ETP-RI is successful, it is not easy to reperform the procedure if a new RBO develops.

This study found that EUS-RI was safely performed in cases where ETP-RI was difficult after multiple SEMS placement for MHBO, and the TRBO of EUS-RI was longer and more favorable than that of ETP-RI. EUR-RI has the potential to become the first choice of RI in the future depending on the progress of devices and technologies. Because this was a retrospective study, future randomized controlled trials comparing EUS-RI with ETP-RI are needed to investigate the tolerability and efficacy of EUS-RI.

In conclusion, as a method of RI after multiple SEMS placement for MHBO, EUS-RI is useful as a rescue technique in cases where ETP-RI is difficult.