Lethal post-endoscopic retrograde cholangiopancreatography pancreatitis following fully covered metal stent placement in distal biliary obstruction due to unresectable cholangiocarcinoma

Authors


Corresponding: Takao Itoi, Department of Gastroenterology and Hepatology, Tokyo Medical University, 6-7-1 Nishishinjuku, Shinjuku-ku, Tokyo 160-0023, Japan. Email: itoi@tokyo-med.ac.jp

Abstract

Biliary self-expandable metallic stent (SEMS) is the preferred and first-line therapy for unresectable malignant biliary obstruction. To date, several reports have revealed the relatively high incidence of acute complications such as pancreatitis and cholecystitis due to mechanical stent compression. In the present case, we encountered fatal pancreatitis following fully covered metal stent placement. An 85-year-old man had middle bile duct strictures due to cholangiocarcinoma. A 10-mm diameter fully covered SEMS was placed across the papilla for biliary decompression. Laboratory data and physical evidence the following day revealed acute pancreatitis. Therefore, antibiotics and protein degeneration enzyme inhibitors were given. However, hiscondition did not improve. Two days after the procedure, we removed the stent and returned him to his original hospital. Serum amylase level decreased below 400 mg/dL 6 h after the procedure. However, the acute pancreatitis worsened. Although we treated the patient in the intensive care unit, he died 32 days after the second admission.

Introduction

Endoscopic retrograde cholangiopancreatography (ERCP) with biliary self-expandable metallic stent (SEMS) is the preferred and first-line therapy for unresectable malignant biliary obstruction.[1] Large-bore metal stents allow for excellent drainage and long stent patency, compared with traditional plastic stents. Therefore, although metal stents are more expensive than plastic stents, they are thought to be cost-effective as a terminal care approach for patients with unresectable disease. In contrast, potential disadvantages of metal stents are rare but possible serious complications, such as cholangitis, pancreatitis and cholecystitis. Furthermore, once serious complications occur, reintervention, the removal or exchange of stents, is more difficult or impossible for metal stents compared with plastic stents. Herein, we describe a case of lethal post-ERCP pancreatitis following fully covered metal stent placement for biliary obstruction due to unresectable cholangiocarcinoma.

Case Report

An 85-year-old man was transferred to our hospital from another hospital with obstructive jaundice due to cholangiocarcinoma. Percutaneous transhepatic biliary drainage (PTBD) was done at another hospital. Endoscopic SEMS placement was conducted. Duodenoscope showed a normal orifice (Fig. 1). As transpapillary biliary deep cannulation was difficult and a small amount of contrast medium went into the pancreatic duct, we used the ‘rendezvous’ technique through the PTBD route for bile duct access. Cholangiography showed the middle bile duct stricture and a relatively narrow intrapancreatic duct (Fig. 2). Endoscopic sphincterotomy was carried out without bleeding or perforation using an automatic cut-out system (Endocut mode, ICC 200; Erbe Elektromedizin, Tübingen, Germany) (Fig. 3). Finally, a 10-mm diameter and 6-cm long fully covered SEMS (WallFlex; Boston Scientific Japan, Tokyo, Japan) was placed across the papilla (Fig. 4). After stent placement, the patient returned to the previous hospital on the same day. He had abdominal pain 6 h after the procedure. Laboratory data showed 13 × 109/L leukocytes and hyperamylasemia (1200 mg/dL). Antibiotics and protein degeneration enzyme inhibitors were given for post-ERCP pancreatitis. However, his condition did not improve. Two days after the procedure, we were consulted by his doctor as well as other doctors, and he was transferred to our hospital again 3 days after the initial procedure. Computed tomography (CT) showed severe acute pancreatitis (Fig. 5). Laboratory data still showed hyperamylasemia (1440 mg/dL). As we surmised that possible obstruction of the pancreatic duct was caused by the fully covered metal stent, we tried to remove the stent after obtaining informed consent from the patient and his family. Stent removal was done with ease using snare forceps (Fig. 6a). Surprisingly, we observed pus emerging from the pancreatic duct during the procedure (Fig. 6b). We then confirmed that there was a possibility of obstruction of the pancreatic duct due to fully covered metal stent compression. Eventually, a 7-Fr plastic stent was placed for biliary drainage. Six hours after the procedure, the serum amylase level decreased below 400 mg/dL. However, the acute pancreatitis worsened. Although we treated the patient in the intensive care unit, he died 32 days after the second admission.

Figure 1.

Endoscopic finding of native papilla.

Figure 2.

Endoscopic cholangiography shows middle bile duct stricture and a relatively narrow intrapancreatic bile duct.

Figure 3.

Endoscopic sphincterotomy was carried out.

Figure 4.

Placement of fully covered metal stent. (a) Fluoroscopic imaging. (b) Endoscopic imaging.

Figure 5.

Computed tomography shows severe acute pancreatitis.

Figure 6.

Stent removal using snare forceps. (a) Snare forceps grasping the metal stent. (b) Pus oozing from the pancreatic duct during stent removal.

Discussion

In the present case, there are several critical issues. First, the cause of the pancreatitis is the most debatable event. To date, a relatively high incidence of post-metal stent placement pancreatitis has been reported of up to 7.3%.[2-7] Several endoscopists have suggested that transpapillary SEMS placement could be one of the risk factors.[5, 8-10] Shimizu et al. revealed that in these pancreatitis cases, non-pancreatic cancer and contrast injection into the pancreatic duct were predictive risk factors.[10] In the present case, the patient had a cholangiocarcinoma, but cancer of the head of the pancreas indicates a non-obstructive pancreatic duct and normal pancreatic juice flow. Theoretically, there are two major mechanisms of post-metal stent pancreatitis, as follows: (i) pancreatic duct opacification; namely, common post-ERCP pancreatitis due to the ERCP procedure; and (ii) radial expansion forces of the SEMS. In the present case, a small amount of contrast medium was injected into the pancreatic duct; therefore, as a cause of pancreatitis, pancreatic duct opacification cannot be entirely eliminated. Nevertheless, the fact that removal of the stent improved the serum hyperamylasemia and that pus could be observed from the pancreatic duct orifice may mean that the radial expansion forces of the SEMS caused the pancreatitis. In particular, the patient did not have a large papilla and a distal bile duct, so a 10-mm diameter SEMS was used in this case. This means that the SEMS over-dilated the papilla even after endoscopic sphincterotomy, causing the SEMS to compress the pancreatic duct orifice. In addition, WallFlex has a distal flare, which means there is a stronger compression towards the papilla when the distal flare is located at the papilla. Although a recent meta-analysis revealed that there was no difference in the rate of pancreatitis between covered and uncovered metal stents,[11] care should be taken, because, theoretically, a covered distal flare allows possible obstruction of the pancreatic duct orifice. However, if pancreatic duct obstruction was absent, as in cancer of the head of the pancreas, it seems that the possibility of the incidence of pancreatitis due to metal stent compression should be comparatively low. Therefore, as a recent study[10] has suggested, a non-obstructive pancreatic duct should be paid attention to if SEMS placement is carried out in distal biliary obstruction.

Second, in the present case, the patient eventually died due to multi-organ failure caused by severe pancreatitis. Not only the severity of the pancreatitis, but old age (85 years), may have caused the patient's death. Nevertheless, based on the clinical course and our speculation, if we could have removed the metal stent on the same day as stent placement, the severity of pancreatitis might not have increased. From this aspect, if we encounter a case similar to the present case, immediate stent removal would be preferred. Recently, a stent removal technique has been reported[11-14] in which the results showed high success rates with a low complication rate. In their review, Ishii et al. described that braided and covered metal stents could be removed easily compared with laser-cut and uncovered metal stents.[14] In the present case, the metal stent was a braided and fully covered type, enabling easy removal.

There are several limitations in the present case. This is only one case in which we were able to observe evidence of an obstruction of the pancreatic duct due to a metal stent. As Adler and Baron mentioned,[15] as they could not find any cause of metal stent placement-induced pancreatitis based on their abundant case series,[16] large and prospective trials are warranted to address this question more definitively.

In conclusion, we encountered fatal pancreatitis following fully covered metal stent placement. Although further studies are mandatory, our case may suggest that endoscopists should pay attention to fully covered metal stent placement in patients with non-pancreatic cancer and pancreatic duct opacification.

Acknowledgments

The authors are indebted to Maya Vardaman, Associate Professor Edward F. Barroga and Professor J. Patrick Barron, Chairman of the Department of International Medical Communications of Tokyo Medical University, for their editorial review of this manuscript.

Conflict of Interests

Authors declare no conflict of interests for this article.

Ancillary