EUS directed enteroenterostomy and endoscopic subucosal dissection for a refractory hepatic jejunal anastomotic stricture in a liver transplant recipient. This case involves a 55-year-old male with a past medical history of a remote cholecystectomy complicated by a vascular injury to the hepatic artery requiring an orthotopic liver transplant, or OLT, with a Roux-en-Y hepatic jejunostomy, or HJ. The patient's OLT occurred approximately 25 years before his presentation, and during that time he only had one episode of transplant rejection that occurred one year after his OLT. Of note, the patient had been stable in tacrolimus 2 mg daily since that time without any further episodes of rejection. This image illustrates the patient's anatomy at presentation, post-OLT and Roux-en-Y HJ. Here you can see the stomach labeled as S, the Roux-en-Y connection between the distal duodenum and jejunum labeled with R, the HJ connection labeled with H, the transplanted liver labeled with an L, and the biliary tree that drains into the HJ anastomosis labeled with a B. The patient initially presented in April 2023 with concerns of two months of progressively worsening biliary symptoms. His lab work can be seen below. Of note, the patient's tacrolimus level remained therapeutic during this time. Thus, with concerns for a possible transplant rejection, the patient underwent a liver biopsy in May that showed no evidence of transplant rejection. With progressively worsening symptoms and lab work without a clear radiology, the patient had an MRCP. As we can see, the MRCP showed a stricture at the arrow consistent with an anastomotic stenosis, so the patient was referred to advanced GI for further intervention. At this point, given the patient's significantly altered anatomy, there were four possible interventions that were discussed. First, a laparoscopy-assisted ERCP and percutaneous drainage were discussed, but the patient preferred non-surgical options that had a lower likelihood of re-intervention, so he deferred these options. Next, an enteroscopy-assisted ERCP with either a single or double balloon was an option, but this intervention had previously been attempted six years prior unsuccessfully for similar symptoms, so it was not pursued. At that time, instead, the patient was successfully treated with an EUS-directed transenteric ERCP or EDI procedure, so the patient chose that option again this time. To do this, the scopes needed for the procedure first needed to be able to reach the HJ anastomosis, so a gastrojejunal or GJ anastomosis needed to be created. Here we can see a depiction of this with the arrows showing the path that the GJ stent was placed between from the stomach to the jejunum to allow for the ERC scope to reach the biliary tree. First, imaging was obtained showing the stomach seen with the arrows closely approximated with the jejunum with surgical clips seen at the top and finally a dilated biliary tree. Next, the EUS scope was positioned in the stomach facing the HJ anastomosis to confirm the biliary dilation. As is evident in this image, the upper third of the main bile duct was dilated up to 12 millimeters with proximal dilation seen in the hepatic duct and bifurcation, which can be seen in this image as the V-shaped structure. Thus, the EUS-guided GJ was pursued. To do this, the afferent or jejunal limb was flooded with a mixture of saline and contrast for expansion of the limb and for better visualization. This process can also be seen here under fluoroscopy where the contrast slowly fills the limb. Next, under EUS guidance, a 20 millimeter by 10 millimeter luminoposing metal stent or LAMS was placed from the gastric body into the jejunum with the distal flange seen deploying into the jejunum being pulled back and then the proximal flange being deployed into the stomach. Fluoroscopic visualization of this can be seen here with the LAMS identified at the arrow being dilated after placement. Flow of saline and contrast can be seen from the jejunum into the stomach after stent placement indicating proper placement and an image of the final stent placement can be seen here. After this, the patient was discharged home to allow time for fistulization to occur and the LAMS to expand on its own, reducing the risk of dislodgement and perforation upon dilation of the LAMS. Four weeks after placement of the GJ LAMS, the patient was brought back for his ETI procedure. For this procedure, a pediatric colonoscope was introduced and guided to the GJ LAMS. Next, the scope was advanced through the LAMS into the afferent limb of the jejunum where the site of the HJ was investigated. However, a clear anastomosis could not be identified. Given this, the pediatric colonoscope was removed and a linear therapeutic echoendoscope was inserted. However, it could not be advanced to the appropriate position, so a forward-viewing therapeutic echoendoscope was inserted and threaded to the site of the HJ. Here, we can see an image of a forward-viewing echoendoscope. This scope may allow us to approach the anastomosis site more vertically and shorten the puncture distance, contributing to the feasibility and safety of the procedure. Under EUS, the common hepatic duct was then punctured with a 19-gauge needle and bile was aspirated, confirming the location. Next, contrast was injected and a cholangiogram was obtained. An 8x8 mm LAMS was then placed between the common hepatic duct and the jejunum with the distal flange seen here deploying into the bile duct and the proximal flange deploying into the afferent jejunum with good apposition. One month after the placement of the CDJ LAMS, the patient reported significant improvement in his symptoms and lab work showed significant improvement. This remained stable for approximately three months, at which point he was re-scoped with a standard EGD scope to evaluate the GJ and CDJ stents. Once the CDJ LAMS was visualized, it was removed using a grasping device. Contrast was injected into the biliary tree to create a cholangiogram that appeared to be good, and the anastomosis between the common hepatic duct and jejunum was dilated to a maximum size of 10 mm using a balloon dilator. Over the next two to three months, the patient reported recurrent symptoms to his presentation and lab work worsened, as can be seen below. Given this, an MRCP was pursued to identify the etiology. As we can see in the MRCP, a stenosis was again visualized. On this ERCP, a filling defect was visualized, so the stone was removed. However, a stenosis was re-encountered at the CDJ, so this was dilated to 8 mm and a trans-papillary covered metal stent was placed with bile seen freely flowing through the anastomosis. After this dilation and ERCP, the patient's lab work improved but did not return to his baseline. Given recurrent stricturing at the CDJ fistula in a short time frame, we opted for a repeat ERCP with endoscopic submucosal dissection, or ESD, to disrupt prior fibrosis and recreate a more durable anastomosis. Following stent removal with a rat tooth forceps, sludge was swept from the ducts and an endoscopic submucosal dissection, or ESD, for tissue destruction and stricturoplasty of the stenosis CDJ was pursued with an IT knife. As we can see, initially the IT knife was used to remove tissue from the most distal portion of the anastomosis. These first cuts were used to determine the extent of the fibrosis, and then as the fistula became wider, the knife was inserted further into the fistula and layers of fibrotic tissue were peeled away to the depth of the biliary tree. In this process, we subsequently encountered suture material that we removed. After this, 2 mL of triamcinolone was injected into the stricture to decrease the risk of fibrotic changes. Finally, a 10 French by 4 cm covered metal biliary stent was placed 3 cm into the common hepatic duct to allow for appropriate remodeling, and bile was seen freely moving through the stent. The patient was discharged home with plans for stent removal in 6 months. To conclude, endoscopic access and cannulation of biliary anastomosis with repetitive treatment can be challenging in patients with altered anatomy due to a liver transplant. An ED-E procedure is an innovative technique that can be used to facilitate access to the pancreatic biliary limb. EUS choledochodruginostomy may be used to successfully treat completely obliterated HD anastomosis. However, given that it is a fistula, there is a high risk of restenosis if the stent is removed. A forward-viewing echoendoscope may be advantageous in this setting given its vertical puncturability and on-foss apposition. ESD and stricturoplasty may be performed using a needle knife to effectively reduce refractory surgical strictures and allow for the recreation of anastomosis. This case demonstrates the ability to combine an endosynography, ERCP, and ESD to successfully manage a complex pathology in a complex surgical anatomy. Further studies are warranted.

liver transplant

biliary stricture

endoscopic procedure

choledochoduodenostomy

endosonography