In this study the investigators retrospectively report outcomes of direct transluminal EUS-BD in a series of patients with malignant biliary obstruction after failed ERCP as the experience of a single Italian center

INTRODUCTION Endoscopic retrograde cholangiopancreatography (ERCP) with placement of biliary stents is the treatment of choice for palliation of malignant obstructive jaundice and has a success rate of 90% with low morbidity rate.1 In 5 to 10% of cases, even in expert hands, stents' placement failed for several reasons as altered anatomy due to surgical intervention, gastric outlet obstruction, ampullary tumors invasion, high grade biliary strictures and all other causes of failed biliary cannulation.

In these unfortunately cases alternative methods have been developed. Percutaneous transhepatic biliary drainage (PTDB) is a efficacy technique but is associated with an adverse events rate of 30% and a negative impact on the quality of life of patients due to the external drainage;4 furthermore surgical biliodigestive anastomosis is burdened by a morbidity and mortality of 30% and 10% respectively.

An effective alternative to PTDB, introduced for the first time in 1996, is endoscopic ultrasonography-guided biliary drainage (EUS-BD). EUS-BD can be performed by four different routes: EUS-guided hepaticogastrostomy, choledochoduodenostomy, rendezvous and anterograde transpapillary drainage.

Among these, rendezvous technique seems to be the safest of all EUS-guided procedure at the expense of a not excellent success rate (from 44% to 80%) and with the limit of the need of a accessible papilla by endoscopy.8 These limitations are overcome by direct transluminal EUS-guided approach as hepaticogastrostomy and choledochoduodenostomy that also ensure a 1-stage procedure.

In this study the investigators retrospectively report outcomes of direct transluminal EUS-BD in a series of patients with malignant biliary obstruction after failed ERCP as the experience of a single Italian center.

Definitions:

Technical success was defined as the correct placement of the metal or plastic stent across the stomach or duodenum to the chosen biliary branch, with radiologically and endoscopically confirmed.

Early clinical success was defined as a drop of bilirubin hematic level by 50 % after 2 week from EUS-BD, while late clinical success was considered as the reaching of hematic bilirubin level compatible with a possible chemotherapy treatment at 3-4 weeks after the endoscopic performance.

Procedure-related adverse events were recorded and graded as mild if they resolved spontaneously, moderate if they required a specific intervention without the need for an extension of hospitalization and severe in case of death or if they required a specific intervention (surgical or not) with consequent prolongation of hospitalization.

Stent patency duration was defined as the time between stent placement and its occlusion Re-stenting was defined as the necessary to second EUS-guided stent placement in patients who didn't achieve early clinical success or in the case of jaundice recurrence from the first treatment.

Inclusion Criteria:

• patients over 18 years old, malignant bile duct obstruction with unsuccessful ERCP drainage.

Exclusion Criteria:

• Patients with benign stricture

All patients afferent to Santa Maria Nuova Hospital in Reggio Emilia, between January 2011 and November 2017, with malignant obstructive jaundice, in whom ERCP had failed