A 74-year-old man had undergone right hepatopancreaticoduodenectomy for bile duct cancer. One year after surgery, obstructive jaundice developed due to a benign anastomotic biliary stricture. Endoscopic retrograde cholangiopancreatography (ERCP) was attempted using an oblique-viewing endoscope, but the anastomosis was not identified. Therefore, endoscopic ultrasonography-guided rendezvous technique (EUS-RV) was performed with the patient’s consent. The intrahepatic bile duct (IHBD) (B2) was punctured from the esophagus with a 19-gauge needle. After injecting the contrast medium, a 0.025-inch guide wire was introduced through the needle. However, the guide wire could not be passed and manipulated through the tight anastomosis, so we changed the 19-gauge needle to an ERCP tapered catheter from the puncture route (Fig. 1). After successfully passing the guide wire through the anastomosis, the oblique-viewing endoscope was replaced with the ultrasound endoscope while keeping the guide wire in place. The guide wire was grasped with a loop cutter and pulled out through the working channel before performing over-the-wire cannulation. The anastomosis was dilated with a 6-mm balloon and two biliary plastic stents were placed through the anastomosis.

After the procedure, the patient had no symptom and came back to his room. However, pneumoderma developed. One hour later after the procedure, computed tomography revealed pneumoderma, mediastinal emphysema, and bilateral pneumothorax (Fig. 2). Chest drainage was performed immediately for the left side on the same day. Antibiotics were given for 5 days. On day 2, the pneumothorax improved. Pneumoderma and mediastinal emphysema gradually improved, too. The patient was discharged home on day 6. After the biliary drainage (BD), obstructive jaundice improved.

ERCP is the standard technique for relief of biliary obstruction. Therapeutic ERCP requires deep cannulation and can be facilitated by the use of guide wire, precutting procedures, and guide wire placement in the pancreatic duct. Although these advanced techniques have increased the success rate of deep cannulation to as high as 97% to 98.5%, the procedure is still not perfect.^1–3 The development of a linear array ultrasound endoscope has enabled various procedures of endoscopic ultrasonography-guided BD (EUS-BD). EUS-BD was reported for the first time in 2001⁴ and is now developing alternative to percutaneous transhepatic BD and surgery in patients in whom ERCP has failed. EUS-BD includes EUS-guided choledochoduodenostomy, EUS-guided hepaticogastrostomy (EUS-HGS), EUS-guided hepaticoenterostomy (EUS-HES), and EUS-RV among others. Since the initial report in 2004,⁵ several studies have supported the use of EUS-RV as an effective salvage technique to achieve biliary cannulation after failed ERCP. The EUS-RV technique comprises three methods based on the approach route: transgastric, transduodenal in a short endoscopic position, and transduodenal in a long endoscopic position. By far, there have been no reported significant differences in rendezvous success and complication rates among these approaches.

The transgastric route for EUS-RV was first described in 2004.⁶ With this technique, the IHBD of B2 or B3 is punctured from the cardia or lesser curvature of the stomach. Compared with the transduodenal approach, the transgastric approach has the major advantages of puncture through the liver parenchyma and less bile leakage. An additional advantage of this route is the straight position of the scope, which is easy to maintain during scope change. However, the route from the transgastric puncture site to the papilla is long and tortuous. Therefore, this technique requires skilled guide wire manipulation. Compared with the B3 IHBD, the B2 IHBD is straighter anatomically and makes guide wire manipulation easier. Before puncturing the B2, the ultrasound endoscope is sometimes positioned in the esophagus. The transesophageal route allows easy puncture of the B2, easy manipulation of the guide wire, and changes of the scope. At our hospital, we used to select this route for EUS-RV in some cases. However, after experiencing this case, we do not recommend the transesophageal procedure in consideration of patient safety.

Notably, a transesophageal approach carries the potential risks of mediastinal emphysema, mediastinitis, and pneumothorax. After experiencing this case, we have made it a practice to clip the esophagogastric junction using a forward-viewing scope before performing EUS-BD through the IHBD. This way, the position of the clip can be easily confirmed under fluoroscopy and puncture through the esophagus can be prevented (Fig. 3).

EUS-RV with the transesophageal approach may have severe complications. These risks further increase during EUS-HGS and EUS-HES than EUS-RV because both procedures necessitate dilation the puncture route. We perform clipping the esophagogastric junction to prevent puncture through the esophagus. In cases of inadvertent transesophageal puncture in spite of the clipping, devices other than the needle should not be passed through the site. We recommend percutaneous transhepatic BD for cases where in the transesophageal route is the only option available during EUS-BD.