1031 EUS-GUIDED INTRAHEPATIC ACCESS FOR RETROGRADE, ANTEGRADE OR TRANSGASTRIC BILIARY DRAINAGE: INDICATIONS, EFFICACY AND SAFETY FROM AN 8-YEAR TERTIARY CENTRE EXPERIENCE.

Abstract

Intrahepatic access for EUS-guided biliary drainage (IH-EBD) still lacks convincing evidence on advantages and risks over comparators. Retrospective series are burdened by small size and heterogeneous inclusion of patients and procedures, therefore IH-EBD struggles to find a role in clinical algorithms. We revised the 8-year experience of a tertiary, academic, referral centre. All consecutive IH-EBDs performed between 2012-2019 were retrospectively included. Indications, technical details, clinical/biochemical follow-up and outcomes were extracted from individual patient files. Variables are reported as proportions or median[interquartile range]. χ-squared, Kruskal-Wallis test and Kaplan-Meier curves were used. 104 IH-EBDs were identified [(malignancies=87 (83.7%); failed ERCP=81 (77.9%); post-surgical anatomy=23 (22.1%)]. Distal, hilar and anastomotic strictures represented 50%, 28.9% and 14.4% of indicons. 16 transhepatic ERCP-rendez-vous (RVs), 43 transhepatic antegrade biliary stentings (ASs) and 45 hepatico-gastrostomies (HGs) were identified. Overall technical success was 89.4%, while clinical success (lowering bilirubin or management of choledocholithiasis) was 94%. Using the ASGE lexicon, overall, severe and fatal complication rates were 16.7%, 3.0% and 0.9% respectively. Median hospital stay was 7 [2-10] days, 4.5 [1-9] in case of no complications. Stent dysfunction occurred in 17.1% after a median of 103.5 [42.5-168.0] days, resulting in a 72% probability of 6-months dysfunction-free survival. Comparing the 3 techniques, benign diseases were more prevalent among RVs (p=0.0004), while hilar/anastomotic strictures were mainly managed through HGs (p<0.0001). Technical failures were higher among RVs (p=0.036). No difference in clinical success or severe adverse events was identified. A lesser extent of bilirubin decrease was noticed among HGs (≥50% decrease in 53.3% vs. 66.7% and 96% of RVs and ASs; p=0.007), which may be attributed to a significantly higher rate of disconnected ducts (53.5% vs. 6.2 and 2.2%, p<0.0001). A trend towards reduced stent dysfunction was noted when HGs were created with half-covered purpose-specific stents (N=17 [38%]) vs. older stents (6.7% vs. 31.6%, p=0.0789). The intrahepatic route for EUS-guided biliary drainage in failed ERCPs or surgically altered anatomy has a good clinical efficacy, a relatively low dysfunction rate and an acceptable safety profile. These results seem especially valuable for indications in which the only alternative would be percutaneous drainage. Increased technical expertise, specifically designed tools and high-quality comparisons are compelling for a standardized inclusion of this technique in the endoscopic armamentarium of tertiary referral centres.