Predictive Factors and Surgical Outcomes of Stent Dysfunction After Preoperative Endoscopic Biliary Stenting in Patients Who Underwent Pancreaticoduodenectomy

Introduction

Malignant biliary obstruction is the most common symptom in patients with periampullary cancer, including pancreatic head ductal adenocarcinoma, bile duct carcinoma, and ampullary adenocarcinoma. Preoperative biliary drainage is often performed for obstructive jaundice which leads to coagulopathy, liver dysfunction, and preoperative cholangitis. However, the necessity of preoperative biliary drainage remains controversial. There are two types of preoperative biliary drainage; percutaneous transhepatic cholangiodrainage (PTCD) and endoscopic biliary drainage consisting of endoscopic biliary stenting (EBS) and nasobiliary drainage (NBD). PTCD is not recommended as the initial preoperative biliary drainage due to the risk of tumor seeding and other adverse events.^1,2 NBD is reported to have a lower incidence of preoperative cholangitis, postoperative pancreatic fistula, and stent dysfunction than EBS.^3–5 On the contrary, NBD has some disadvantages such as nasopharyngeal discomfort, impaired activities of daily living, and malabsorption of fat-soluble vitamins. Therefore, EBS using a plastic stent is currently widely performed for preoperative biliary drainage. The major complication after EBS is stent dysfunction resulting in recurrent jaundice and cholangitis. However, there are few reports regarding the risk factors and surgical outcomes of stent dysfunction after EBS. The aim of this study was to investigate the risk factors and surgical outcomes of stent dysfunction after EBS in patients who underwent pancreaticoduodenectomy (PD).

Patients and Methods

Between November 2008 and January 2019, 94 consecutive patients underwent PD with pancreaticojejunostomy and modified child reconstruction after EBS using a plastic stent at Department of Surgery, in the Jikei University Hospital, Tokyo, Japan. Of these, 9 patients were excluded, 2 without malignancy, 2 with neoadjuvant chemotherapy, 1 with 10 Fr stent at the initial EBS, and 4 due to lack of data, leaving the remaining 85 patients for this study.

Stent dysfunction was defined by insufficient biliary drainage and the presence of various symptoms, including high fever and abdominal pain, with elevated serum hepatobiliary enzyme levels or bilirubin level. For stent dysfunction, stents were replaced by thicker stents or NBD. Postoperative complications such as pancreatic fistula and bile leakage were diagnosed based on the definitions of the International Study Group on Pancreatic Fistula ⁶ and the International Study Group of Liver Surgery. ⁷

At first, we retrospectively investigated the relationship between patient characteristics and the incidence of stent dysfunction using univariate and multivariate analyses. The factors included age, gender, pathological diagnosis (ductal adenocarcinoma, bile duct adenocarcinoma, or ampullary adenocarcinoma), serum total bilirubin (T-Bil) before the initial EBS (≥15 or <15 mg/dL), serum C-reactive protein (CRP) before the initial EBS (≥1 or <1 mg/dL), serum amylase after the initial EBS (≥1000 or <1000 IU/L), and stent diameter (7 or 8.5 Fr).

Next, we analyzed surgical outcomes in relationship to the incidence of stent dysfunction using the following factors: duration from the initial EBS to the operation, operative time, intraoperative blood loss, postoperative pancreatic fistula, and postoperative bile leakage. Also, we compared serum amylase after the initial EBS between 7 and 8.5 Fr stent group.

This study was approved by the Ethics Committee of the Jikei University School of Medicine (30–322).

Results

Discussion

We herein demonstrated that T-Bil before the initial EBS ≥15 mg/dL and 7 Fr stent was significant predictors of stent dysfunction by univariate analysis, and 7 Fr stent was the only significant independent risk factor for stent dysfunction after EBS by multivariate analysis. Hashimoto et al. reported that a male gender, a stent diameter ≤8 Fr, and an endoscopic retrograde cholangiopancreatography (ERCP) procedure time ≥45 minutes were the risk factors of stent dysfunction. ⁸ van Boeckel et al. reported that the plastic stent, stricture severity requiring preceding dilatation, and initial T-Bil >50 μmol/L were associated with a shorter period of stent patency. ⁹ Sugiyama et al. reported that tube dysfunction was observed significantly more frequently in patients with pancreatic cancer than in bile duct or ampullary cancer. ¹⁰

Several studies have shown surgical outcomes of stent dysfunction, including postoperative complications and long-term prognosis. Kaneko et al. reported that patients with preoperative cholangitis due to stent dysfunction had a significantly higher incidence of pancreatic fistula than those without preoperative cholangitis. ¹¹ Kurahara et al. reported that preoperative biliary-related inflammation, including tube-obstructive cholangitis and post-ERCP pancreatitis, was a significant risk factor of postoperative complications and a poor predictor of shorter progression-free survival and overall survival in patients with biliary tract cancer. ¹² Although our current study showed that operation time, intraoperative blood loss, postoperative pancreatic fistula, and bile leakage were comparable between the patients with and without stent dysfunction, the number of ERCP in the patients with stent dysfunction naturally significantly more than that in the patients without sent dysfunction group (2.7 ± 0.9 versus 1.2 ± 0.4; P < .0001).

Post-ERCP pancreatitis is one of the most important complications associated with ERCP-related procedures. Several risk factors for post-ERCP pancreatitis have been reported, such as sphincter of Oddi dysfunction, a female gender, a history of pancreatitis, nonplacement of a pancreatic duct stent, procedure time of 30 minutes or more, sampling of pancreatic tissue, and intraductal ultrasonography.^13,14 A mechanism of pancreatitis caused by ERCP is pancreatic intraductal pressure, ¹⁵ which suggests that large-bore stent could cause pancreatitis. The present study showed that 8.5 Fr stent was better than 7 Fr stent in terms of the incidence of stent dysfunction, and serum amylase level after the initial EBS between the patients with 7 and 8.5 Fr stent was comparable. Therefore, for a lower incidence of stent dysfunction, an 8.5 Fr stent could be placed as the initial EBS without increasing pancreatitis.

The current study had some limitations. First, it used a retrospective design and was performed at a single center. Second, the sample size was not large enough. A unique strong point, however, is that this was the first report that analyzed both risk factors and surgical outcomes of stent dysfunction after EBS in patients who underwent PD. We believe that this study provides useful information for the management of patients who undergo PD after EBS.

In conclusion, a 7 Fr stent was the significant independent predictor of stent dysfunction after EBS in patients who underwent PD. An 8.5 Fr stent should be considered as the initial EBS.