Complex biliary intervention: Percutaneous small bowel access confirmation with cone-beam computed tomography and retrograde biliary obstruction recanalization

Abstract

Introduction

Patients with central biliary obstructions are frequently managed by endoscopic intervention, but in patients who have undergone hepaticojejunostomy (HJ) or Roux-en-Y gastric bypass, the surgically altered anatomy can preclude endoscopic treatment for biliary obstruction.1 Interventional radiologists still play an important role in these cases as they can access the biliary system percutaneously. Many biliary obstructions can be treated with percutaneous transhepatic biliary drainage (PTBD), but some biliary obstructions are not easily crossed with typical antegrade recanalization.1 In cases of obstruction with multiple isolated ducts, routine PTBD would require multiple percutaneous drains to effectively drain the biliary system. Some of these cases may be better served by retrograde biliary recanalization after obtaining percutaneous jejunal access.

The use of a jejunostomy access to perform biliary interventions is an established technique for the treatment of biliary stenosis in patients with HJ anastomosis.1–4 Percutaneous small bowel access can be a challenging procedure if the intestinal loop is not fixed to the anterior abdomen with metallic markers usually placed during the initial surgery.1,2 Fluoroscopic aids in locating a target loop of bowel such as iodinated contrast and air can quickly move out of the bowel, and do not exclude an interposed segment of colon between the percutaneous access site and the target loop of small bowel. Thus, bowel fixation and dilation are necessary for percutaneous jejunostomy. Cone-beam computed tomography (CBCT) is a technology that has grown in availability in the angiography suite as a means for guiding procedures.5,6 Prior studies have described the safety and utility of CBCT during percutaneous gastrostomy tube placement.5

The purpose of this study is to evaluate the technical usefulness of CBCT in a series of 9 patients where CBCT was successfully used to plan and obtain retrograde small bowel access into the biliary system. In 3 patients CBCT was used to guide creation of a new bilio-enteric anastomosis. In 6 patients CBCT was used to confirm appropriate needle position in loops of bowel not previously surgically fixed to the anterior abdominal wall for retrograde trans-jejunal biliary access.

Methods

Between January 2012 and February 2016 at the University Hospital Medical Center (San Antonio, TX, USA) a retrospective analysis of medical records and images was performed looking for patients who had undergone complex biliary recanalization. Patients were included if the biliary recanalization was performed through or assisted by percutaneous small bowel access formation, and CBCT was used during the procedure. Nine patients who underwent such biliary recanalization procedures were identified via chart review. In all patients, prior attempts for routine ante-grade PTBD placement had provided inadequate biliary drainage, were not attempted due to biliary anatomy, or were not tolerated by the patient. Each patient had a CBCT scan performed during their initial recanalization procedure in order to identify the exact position of jejunal loop and plan a retrograde access. In patients with disconnected biliary systems, CBCT was done to plan a retrograde and antegrade recanalization. Institutional Review Board exemption was obtained for this retrospective review. Procedures were performed in either of two angiography suites equipped with Phillips Allura FD20 angiography units with XperCT capabilities (Philips Medical Systems, Andover, MA, USA). Written informed consent was obtained from all patients. Five patients had their procedures performed with moderate conscious sedation consisting of midazolam and fentanyl, and four patients had procedures performed under general anesthesia provided by the anesthesia service. All patients were provided antibiotic prophylaxis with either ciprofloxacin 400 mg or piperacillin/tazobactam 3.375 g intravenously before the procedure. CBCT was performed using the XperCT program with a standard abdominal setting without intravenous contrast. A dedicated workstation was used for image generation, three-dimensional reconstructions, and intra-procedural image evaluation.

In three patients with disconnected biliary systems, a 21 G Chiba needle (AccuStick System; Boston Scientific, Natick, MA, USA) was advanced through the anterior abdominal wall, both walls of a target loop of small bowel, and into a 9 to 15 mm snare (En Snare; Merit Medical Systems, South Jordan, UT, USA) placed in the biliary system using the existing previously placed PTBD access. After the initial puncture, CBCT was used to confirm appropriate needle position through and through the small bowel into the snare, and that the needle had not passed through non-target loops of small bowel or colon. In two patients the duodenum was first inflated with air using a nasogastric angiographic catheter that was placed in the stomach. After CBCT confirmation of adequate needle position, a 0.018″ guide wire (AccuStick System) was advanced through the Chiba needle and removed through the existing PTBD access with the snare. The new bilio-enteric anastomosis was first dilated with a 4 mm balloon followed by placement of a 10 Fr internal/external biliary drain (Mac-Loc biliary drain; Cook Medical, Bloomington, IN, USA) (Fig. 1).

Figure 1

Patient disconnected biliary system and permanent external biliary drain, multiple failed attempts to cross the occlusion. (A) Cholangiogram demonstrating a blind-ending, obstructed biliary system (arrow). (B) Cone-beam computed tomography showing ...

In six patients with prior HJ and complex biliary stenoses, percutaneous access was obtained into the jejunum with a 21 G Chiba needle, and secured with a 0.018″ wire. In four patients with prior PTBD, the jejunum was opacified and distended with contrast. In the other two patients, the position of the jejunum was identified with CBCT. The jejunum was then fixed to the abdominal wall using two T fasteners (Halyard Health, Alpharetta, GA, USA). Then 6 to 8 Fr sheaths were placed. Through the sheaths a combination of wires and catheters were used to recanalize the biliary occlusions in a retrograde fashion using angled tip catheters and glidewires (Terumo Medical, Somerset, NJ, USA). Following wire recanalization, the obstructing strictures were dilated with 4 to 6 mm balloons. After balloon dilation, internal 10 to 12 Fr plastic biliary stents (Miller Double Mushroom; Cook Medical) were placed to maintain patency. For patients with multiple isolated bile ducts, at least two biliary ducts were accessed and stented. Following internal stent placement, a jejunostomy tube was left in place with 8 to 12 Fr drains to allow future internal stent exchanges. Median patient radiation exposure for the procedures including total fluoroscopic and CBCT dosages were measured using the air kerma method. Technical success was defined as the ability to obtain retrograde biliary access through a small bowel loop.

After the procedure, patients who received moderate conscious sedation were brought to a radiology recovery area and monitored. If the procedure was performed with general anesthesia the patients were brought to the post-anesthesia care unit following the procedure. The decision to admit the patient for overnight observation versus same day discharge was made on an individualized basis at the interventional radiologist’s discretion

In patients with prior PTBD catheters, the biliary tubes were removed after the biliary recanalization and stent placement. Future management of internal biliary stents was via retrograde exchanges. The existing jejunostomy tube was exchanged for a vascular sheath. This was followed by stent removal with snare or forceps. After stent removal the bile ducts were re-accessed with a wire, new internal stents placed, and the sheath exchanged for a jejunostomy tube. Follow-up consisted in cholangiograms with routine catheter and/or stent exchanges every 3 months. Clinical examination and laboratory evaluation of liver function tests were performed during each radiologic visit.

Results

Jejunostomy access

In 6 patients with HJ and biliary obstructions, CBCT was used to assist with the percutaneous jejunostomy creation in order to recanalize biliary occlusions and place multiple internal biliary stents, which were then managed with percutaneous retrograde exchanges. An average of 2.8 plastic retrograde stents was placed per patient (range, 2–4 stents) for a mean time of 6 months (range, 3–11 months). Five patients are currently tube and stent free with no evidence of recurrent biliary obstruction. At 8 months after the initial recanalization procedure, the jejunostomy access in patient 4 migrated out of the jejunum. This was replaced via a new puncture, but the new jejunal access crossed two loops of bowel and created a small bowel obstruction. The crossing of the jejunum was only seen on CBCT after multiplanar reconstruction (Fig. 2). The patient required segmental small bowel resection and has since recovered. The patient had the jejunostomy tube and biliary stents replaced following the small bowel resection. Patient 6 developed cellulitis around the jejunostomy tube access site 2 months following initial placement that was successfully treated with oral antibiotics. Patient 7 had a history of biliary atresia and multiple episodes of cholangitis prior to biliary drainage. The patient had an additional bout of cholangitis after drainage, although this developed greater than 30 days after the internal biliary stent placement. Patient ultimately underwent orthotopic liver transplantation 11 months after the procedure.

Figure 2

Patient with history of orthotopic liver transplant with complex biliary obstruction. Previous right percutaneous transhepatic biliary drainage and left external biliary drain in place. (A) Radiograph shows a sheath placed ...

Discussion

Biliary intervention using transjejunal access has been previously described for patients with HJ and Roux-en-Y.1,2 The multiple advantages of the jejunal access have also been well described including less discomfort than the PTBD access, capability to reach multiple bile ducts with a single puncture and possibility of repeat future interventions using the same access.1 In the largest previously described series, overall procedure-related complication rate of percutaneous transjejunal biliary intervention was 3% including sepsis (most common complication), roux limb perforation, wound infection, portal vein puncture with hemorrhage, and erythematous reaction.1 In the majority of the described cases, the jejunal access is fixed to the skin with surgical markers placed during the initial surgery. Accessing non-marked loops has been described previously using percutaneous transhepatic cholangiography to opacify the loops2,7 or previous CT images to identify the loop. What is common in this series of complex biliary cases is that none of the patients had jejunal access fixed to the skin with surgical markers at the time of initial percutaneous small bowel access, and CBCT was used after small bowel puncture with a 21 G needle to confirm an appropriate access site in these non-fixed loops of small bowel. CBCT is a useful adjunct for percutaneous jejunal access since colon, liver, or other loops of small bowel can be interposed between the abdominal wall and target loops of jejunum. CBCT helps with access site selection by verifying, with cross-sectional imaging, the accurate initial needle placement. In this series, retrograde transjejunal interventions were successfully performed in 6 patients with complex biliary stenoses allowing placement and exchanges of multiple plastic biliary stents. Five patients were successfully treated with satisfactory outcomes, while one patient with multiple intrahepatic strictures and diffuse hepatholithiasis, eventually required orthotopic liver transplant. In this series although the initial technical success was 100%, small bowel obstruction occurred in one patient (Patient 4) resulting from accidental crossing of a jejunal loop with the tube. In this patient, the original jejunostomy tract matured and fixed the loop of small bowel to the abdominal wall. This access was later lost due to catheter migration, and attempts to recanalize the original tract were unsuccessful. A new puncture was performed slightly superior to the original tract. In this case, the previously accessed loop of bowel was accidentally crossed since it was adherent to the anterior abdominal wall. This could be seen on the CBCT images obtained at the time of performing the second jejunostomy access (Fig. 2), but was only noted in retrospect on closer review of multiplanar reconstructions.

Patients with disconnected biliary systems pose a challenging problem with very limited non-surgical options. The constant absence of bile in the small intestine in these patients can lead to severe malnutrition. Previously described non-surgical options include some reports of magnetic anastomosis8 and sharp needle recanalization using an antegrade or retrograde approach. Placement of magnets requires having endoscopic access to small bowel, is technically complex,8,9 and requires serial dilatation of the PTBD tract up to 18 Fr.8 Sharp needle recanalization using a Colapinto needle has been reported in several cases, and in these jejunal access was usually obtained and the puncture done blindly, or by placing an inflated balloon that was used as a target.10–14 In a few cases report, a radiofrequency wire was used in recanalization of complete obstruction where conventional methods had failed in five patients. This is an alternative method for patients with biliary obstructions and isolated ducts, but the radiofrequency wire is not readily available and can be problematic with false lumen passages.14 Recently, Mansueto et al15 reported on their experience in five patients using percutaneous jejunal access to create biliary neoanastomoses in patients with HJ dehiscences and biliary leaks. They used a similar technique to that used in 3 patients of this series by creating new percutaneous bilio-enteric anastomoses, although with some differences: they did not use CBCT, and they used sharp recanalization with a Rosch–Uchida transjugular liver access set needle (Cook Medical) to puncture an already opacified biliary system. In this series of 3 patients with disconnected biliary systems, percutaneous small bowel access was used as an adjunct to help with recanalization and formation of new bilio-enteric anastomoses. The axial images obtained with the possibility of obtaining multiplanar reconstructions, were very useful to obtain a safe access through the two walls of the target small bowel loop into the biliary system. Two patients with new bilio-enteric anastomosis created had long-term successful outcomes; while in a third patient, the anastomosis reoccluded eventually requiring surgical revision.

In conclusion we are reporting a series of 9 patients were CBCT was successfully used to plan and obtain retrograde small bowel access into the biliary system. In three patients, CBCT was used to guide the creation of neo bilio-enteric anastomosis; while in 6 patients, CBCT was used to confirm a secure access into jejunum loops that were not surgically fixed or marked. CBCT can be a helpful adjunct technique when performing complex biliary procedures. It can be used before percutaneous puncture, if desired, to help identify safe windows for percutaneous small bowel access. It can also be used after the initial puncture to confirm that the needle has not traversed colon or other non-target structures in a through and through fashion. If colon or other non-target structures have been traversed, CBCT allows the opportunity to identify this and correct the needle placement without increasing the size of the hole created in those structures. The main drawbacks of using CBCT are increased procedural time and increased radiation dose, so selective use of CBCT should be employed to minimize these. Limitations of this study include its retrospective nature and a limited number of cases. Further studies are needed to validate the safety and utility of CBCT to plan complex biliary procedures.

Article information

References