Afferent limb syndrome, biliocutaneous fistula, bilioentrocutaneous fistula, dehiscence of the hepatic or jejunostomy. Can we treat all of this endoscopically? A 53-year-old female with multiple surgical necrosectomies, Whipple in 2002 for necrotizing pancreatitis, including entroenterostomies, was referred to us with recurrent cholangitis and leakage around a long-standing percutaneous biliary drain, likely due to an associated afferent limb syndrome. CT scans showed dilated intrahepatic ducts with two existing percutaneous biliary drains going through a central abdominal incisional scar through the left intrahepatic ducts into the afferent jejunal limb. Prior failed attempts at treating this included double balloony ERCP due to severe adhesions and stricturing of the afferent limb with failure to reach the hepatic or jejunostomy. So she was taken for a hepatic or gastrostomy. Using Doppler to avoid vessels, the left intrahepatic duct was punctured with a 19-gauge needle. A cholangiogram was obtained alongside the existing drain. Over an angled 0.025-inch guide wire, a 4-millimeter balloon dilation was performed. An 8-millimeter by 8-centimeter fully covered biliary metal stent was deployed at the EG junction, through which a 7-French by 15-centimeter plastic stent was placed into the right intrahepatic duct. Despite the hepatic or gastrostomy, over the next two weeks the patient continued to have a high output from the biliocutaneous fistula. The suspected reason was the leak was not adequately decompressed by the hepatic or gastrostomy alone. So she was taken for another procedure to treat the downstream stricture in the afferent jejunal limb to see if better drainage would help. Through the hepatic or gastrostomy, an extraction balloon was passed through the biliary tree into the afferent limb, where a high-grade stricture was identified. This was dilated with a 10- to 12-millimeter balloon, followed by the placement of two pigtail stents across the stricture to provide even better downstream biliary drainage. Over the next two weeks, the output from the fistula diminished but did not resolve. The suspected reason was the leaking segment was not being adequately decompressed by the current hepatic or gastrostomy. So she was taken for a third procedure with the intention to place a new hepatic or gastrostomy into the branch of the bile duct with the long-standing PTBD to better drain it. The previously placed hepatic or gastrostomy at the EG junction was removed. Contrast injection alone did not help localizing this segment on EUS. So a 9-millimeter balloon was passed percutaneously into the leaking segment and inflated with contrast. Now this was identified with EUS, but in a very difficult scope position. And a 19-gauge needle was used to puncture the balloon and gain access to this leaking segment. A new guide wire was passed into this segment, which was grasped by IR and removed percutaneously. Scope position was very challenging, precluding complete tract dilation using a balloon via EUS, despite the externally secured wire. So a 7-French stent retrieval device was used to dilate the tract along with a 4-7-French dilating catheter passed percutaneously. A new hepatic or gastrostomy was then created using a 10-millimeter by 8-centimeter fully covered biliary metal stent to cover the leaking segment. Over the next four weeks, the output from the biliocutaneous fistula decreased but did not resolve completely. The suspected reason was the chronicity of the fistula going through a central scar and not the abdominal wall musculature. So the patient was taken back for a fourth procedure. Cholangioscopy via the hepatic or gastrostomy was performed, demonstrating food material in the biliary tree. And the site of the biliocutaneous fistula was identified. A guide wire was passed percutaneously to confirm this tract. The fistula tract was then ablated using an APC probe passed percutaneously and visualized cholangioscopically. A nasobiliary drain was also placed via the hepatic or gastrostomy to provide intermittent suctioning to further aid healing of this fistula. Nasobiliary drain to low intermittent wall suction. Over the next two weeks, the existing biliocutaneous fistula closed and nasobiliary drain was removed. However, 12 weeks later, she presented again with a new fistula, this time with bile and food material emerging, consistent with a bilioentrocutaneous fistula. Cholangioscopy via the hepatic or gastrostomy was repeated. Now the pathology for the new entrocutaneous fistula was evident. There was an anastomotic breakdown at the surgical hepatic or jejunostomy with a large abscess containing food material. This was meticulously cleaned out with a rat-toothed forcep, either out through the bile ducts or passed downstream into the afferent jejunal limb. View once more of the bile duct approaching the surgical hepatic or jejunostomy adhesions and abscess, which has now been completely cleaned out. Now to treat this complex bilioentrocutaneous fistula, over a guide wire, a 16 French nasogastric tube was passed through the hepatic or gastrostomy into the afferent jejunal limb and placed to suction. Two weeks later, there was complete closure of the bilioentrocutaneous fistula and NG tube was removed. Two-year follow-up. She has been on a low-residue diet with no recurrent cholangitis, abscess, or fistula. She has had three procedures to change out the pigtail biliary stents for intermittent pain and remains a non-surgical candidate. Clinical implications and techniques to highlight. Understand the anatomic problem driving the pathology. In this case, afferent limb syndrome, chronic percutaneous biliary drain placed through a ventral abdominal scar, and an anastomotic breakdown at the hepatic or jejunostomy. When afferent limb pigtail stents fail and an EUS hepatic or gastrostomy are also insufficient to resolve a leak, a targeted hepatic or gastrostomy into the leaking biliary segment using IR rendezvous can help. Rarely, covering a fistula even with a large metal stent may not be enough to allow healing. Ablation of a chronic tract and nasobiliary suction can further aid healing of a refractory biliocutaneous fistula. Dehescence and abscess formation at a surgical hepatic or jejunostomy can also be potentially healed by meticulously cleaning out the cavity and using large-caliber drains such as NG tubes to provide suctioning. In conclusion, afferent limb syndrome, biliocutaneous fistula, bilioentrocutaneous fistula, dehescence of hepatic or jejunostomy can be treated endoscopically in the non-surgical patient.

53-year-old female

multiple surgical procedures

recurrent cholangitis

percutaneous biliary drain

afferent limb syndrome