Self-expandable metal vs. plastic stents for preoperative biliary drainage in patients receiving neoadjuvant chemotherapy

Article information

Clin Endosc. 2025;58(6):817-825

Publication date (electronic) : 2025 July 29

doi : https://doi.org/10.5946/ce.2025.045

Takashi Tamura, Reiko Ashida, Yuki Kawaji, Masahiro Itonaga, Yasunobu Yamashita, Masayuki Kitano

Second Department of Internal Medicine, Wakayama Medical University, Wakayama, Japan

Correspondence: Takashi Tamura Second Department of Internal Medicine, Wakayama Medical University, 811-1, Kimiidera, Wakayama city, Wakayama 641-8509, Japan E-mail: ttakashi@wakayama-med.ac.jp

Received 2025 February 12; Revised 2025 May 1; Accepted 2025 May 2.

Abstract

INTRODUCTION

Pancreatic cancer has a poor prognosis, with a 5-year survival rate of only 6%.1,2 Pancreatectomy is the only curative treatment for pancreatic cancer. However, many patients have unresectable pancreatic cancer at the time of diagnosis. Only 15% to 20% of patients are candidates for surgical resection.3 Regarding the location, 78% of pancreatic cancers occur in the pancreatic head, and 22% occur in the pancreatic body and tail.4 Cancer of the pancreatic head causes obstructive jaundice and hepatic dysfunction owing to biliary obstruction.5 Jaundice and liver dysfunction can prevent surgery and often necessitate biliary drainage before pancreatic resection.6 Biliary drainage can be achieved using three main approaches: endoscopic retrograde biliary drainage (ERBD), endoscopic ultrasound-guided biliary drainage, and percutaneous transhepatic biliary drainage. ERBD is the most common approach for preoperative biliary drainage (PBD). ERBD was first introduced in the 1970s with the advent of endoscopic retrograde cholangiopancreatography (ERCP), which served as a diagnostic procedure and a therapeutic technique for stent placement in patients with pancreaticobiliary disease.7 The use of plastic stents (PSs) in ERBD was first reported in the 1980s.8 In 1989, self-expandable metal stents (SEMSs) were first used in ERBD for malignant biliary obstruction (MBO) and were subsequently utilized primarily for MBO.9 Previously, SEMSs were used in ERBD only for unresectable MBO, whereas PSs were used in ERBD for benign biliary obstruction or preoperative MBO. Recently, neoadjuvant chemotherapy (NAC) was reported to lead to better postoperative outcomes in patients with resectable and borderline resectable pancreatic cancer.10,11 Therefore, biliary stents capable of sustaining long-term PBD are becoming increasingly desirable. The usefulness of SEMS placement for PBD in patients with pancreatic cancer has been reported, leading to an increased use of SEMSs in ERBD.6 However, there are concerns regarding the increased adverse events associated with the placement of metal stents for PBD and their potential impact on pancreatectomy. This review summarizes the current knowledge on the advantages and disadvantages of SEMSs and PSs for PBD in patients with pancreatic cancer receiving NAC.

DIFFERENCES BETWEEN SEMSS AND PSS IN ERBD

PSs with lumen sizes ranging from 7 Fr (2.3 mm) to 10 Fr (3.3 mm) are commonly used for ERBD in patients with distal MBO. By contrast, SEMSs with a lumen size of 10 mm are often used for ERBD in patients with distal MBO. The lumen size of PSs is smaller than that of SEMSs. A meta-analysis reported that the time to recurrent biliary obstruction (TRBO) was 4.45 months longer with SEMSs than with PSs in patients with unresectable distal MBO.12 The median TRBO of PS ranges from 2.0 months to 7.6 months. In contrast, the median TRBO of SEMS ranges from 3.8 months to 8.9 months.12 However, SEMSs are more expensive per unit than PSs. In terms of stent replacement, PSs are easier to replace, whereas uncovered SEMSs (UCSEMSs) cannot be removed, and covered SEMSs (CSEMSs) cannot be removed once a long period has elapsed after their placement.

NECESSITY OF PBD

No studies have demonstrated a clear benefit of PBD.6 Several retrospective analyses have demonstrated that the overall morbidity does not differ between patients who undergo PBD and those who do not.13-17 Some studies have demonstrated that PBD is associated with prolonged operative duration, greater intraoperative blood loss, and a higher rate of postoperative infections.18-21 In addition, the DROP trial, a multicenter randomized controlled study, evaluated 202 patients with resectable pancreatic head cancer by assigning them to early surgery within 1 week without PBD or ERCP with PBD using a 7-Fr PS, followed by delayed surgery 4 to 6 weeks later. Over 120 days, the cumulative rate of serious adverse events was significantly higher in the PBD group than in the early surgery group (74% vs. 39%, p<0.001), with a relative risk of adverse events in the early surgery group of 0.54 (95% confidence interval [CI], 0.41–0.71). The higher adverse event rate in the PBD group was largely attributed to PBD-related issues, including cholangitis (26%) and stent occlusion (15%), rather than postoperative adverse events related to jaundice. Mortality rate and duration of hospital stay did not differ significantly between the two groups.17

However, early surgery for pancreatic cancer is rare. The use of NAC before surgery for pancreatic cancer has become increasingly common in recent years. Many studies have demonstrated the efficacy of NAC in the treatment of borderline resectable pancreatic cancer.11 NAC is recommended for patients with borderline resectable pancreatic cancer according to the 2021 National Comprehensive Cancer Network guidelines.10,22 Additionally, the application of NAC has been expanded to include patients with resectable pancreatic cancer because it may increase the likelihood of achieving negative surgical margins, target micrometastatic disease, and identify patients who are unlikely to benefit from surgery because of rapid disease progression or previously undetected metastases.11,23 A meta-analysis demonstrated that NAC improved overall survival compared with upfront surgery, with a hazard ratio of 0.66 (95% CI, 0.52–0.85; p=0.001; I²=46%), representing an increase in median overall survival from 19 to 29 months.11 NAC is typically administered over 2 to 6 months prior to surgery.11,24 Moreover, certain types of chemotherapy may require the normalization of liver function before initiation.23,25 Therefore, PBD is required to deliver NAC in patients with MBO-induced liver dysfunction.

SEMSS VS. PSS IN TERMS OF RBO DURING PBD

The prevention of RBO and preoperative cholangitis during PBD is important. If preoperative cholangitis or RBO occurs during NAC, NAC should be discontinued. Furthermore, preoperative cholangitis during PBD increases the incidence of postoperative adverse events such as delayed gastric emptying and is an independent risk factor for severe adverse events following pancreaticoduodenectomy (PD).26

A retrospective analysis of 272 patients with pancreatic cancer undergoing PD compared the perioperative outcomes and adverse events between SEMSs and PSs. A total of 29, 141, and 92 patients had SEMSs, PSs, and no biliary decompression, respectively. Patients with SEMSs had similar perioperative outcomes but a significantly longer time to RBO than patients with PSs (median, 125 vs. 43 days; p<0.001). The patients with PSs experienced more stent-related adverse events and exchanges.27 Therefore, SEMSs have been used for PBD during NAC to reduce the incidence of RBO. Several studies have reported that SEMS placement for PBD in patients with pancreatic cancer receiving NAC is associated with a lower incidence of preoperative RBO than PS placement (Table 1).25,28-32

Table 1.

Summary of studies comparing SEMSs and PSs for preoperative biliary drainage in pancreatic cancer patients receiving NAC

A recent randomized controlled trial involving 22 patients with borderline resectable pancreatic cancer receiving NAC compared the outcomes of ERCP with biliary drainage using a fully covered SEMS (FCSEMS) (Fig. 1) and a 10-Fr PS (Fig. 2). RBO occurred significantly less frequently in the FCSEMS group than in the PS group (18.2% vs. 72.8%, p=0.015). The FCSEMS group also required fewer stent re-intervention procedures (0.27±0.65 vs. 1.27±1.1, p=0.001).25 A cohort of 575 patients who underwent preoperative PBD before PD was analyzed using data from the Dutch Pancreatic Cancer Audit (2017–2018), with 42.8% receiving SEMSs. The use of SEMSs was associated with a lower rate of PBD-related cholangitis, reduced incidence of postoperative pancreatic fistulas, and shorter postoperative hospital stay.33,34

Fig. 1.

Placement of a fully covered self-expandable metal stent (FCSEMS) for biliary obstruction caused by pancreatic cancer. (A) Image of placement of an FCSEMS across a stricture in the bile duct caused by pancreatic cancer. (B) Endoscopic image of the FCSEMS in the papilla after stent placement across the stricture in the bile duct. (C) A computed tomography (CT) scan image showing a mass in the head of the pancreas causing bile duct obstruction before neoadjuvant chemotherapy (NAC). (D) A CT scan image showing that the tumor decreased in size following NAC after FCSEMS placement.

Fig. 2.

Placement of a plastic stent (PS) for biliary obstruction caused by pancreatic cancer. (A) Image of placement of a PS across a stricture in the bile duct caused by pancreatic cancer. (B) Endoscopic image of the PS in the papilla after stent placement across the stricture in the bile duct. (C) A computed tomography (CT) scan image showing a mass in the head of the pancreas causing bile duct obstruction before neoadjuvant chemotherapy (NAC). (D) A CT scan image showing that the tumor decreased in size following NAC after PS placement.

In addition, a meta-analysis showed that the placement of SEMSs for PBD was associated with a lower incidence of RBO than PS placement. In a comparison of stent patency rates at 1, 2, and 3 months post-insertion based on stent type, patients with PSs demonstrated lower patency rates than those with SEMSs (1 month, SEMS [94.0%] vs. PS [75.6%]; p=0.006; 2 months: SEMS [91.1%] vs. PS [48.1%], p<0.001; and 3 months: SEMS [87.0%] vs. PS [26.8%], p<0.001).35 For patients with pancreatic cancer receiving NAC, the duration of PBD often exceeds 2 months, and SEMSs are the preferred choice for PBD. Therefore, SEMSs are considered superior to PSs for PBD in patients with pancreatic cancer receiving NAC because they reduce the risk of RBO.

SEMSS VS. PSS IN TERMS OF ADVERSE EVENTS IN PBD

In PBD for patients with pancreatic cancer receiving NAC, PSs are associated with a higher frequency of RBO than SEMSs, leading to an increased incidence of preoperative cholangitis25 (Fig. 3A, Table 1). SEMS placement is a significant risk factor for post-ERCP pancreatitis and cholecystitis in patients with biliary obstruction caused by unresectable MBO (Fig. 3B, C).36,37 These studies indicate that the rate of post-ERCP pancreatitis was more than five times higher in patients who received SEMSs.37 Furthermore, the incidence of cholecystitis following stent placement was more than three times higher in patients who received SEMSs than in those who received PSs.36 In addition, a meta-analysis reported that post-ERCP pancreatitis occurred more frequently with SEMS than with PS placement for preoperative drainage in patients with pancreatic cancer.35 The incidence of post-ERCP pancreatitis was 10.7% upon placement of SEMSs compared with 4.7% upon placement of PSs (odds ratio [OR], 3.01; 95% CI, 1.48–6.11; p=0.002). Additionally, the incidence of cholecystitis following ERCP was 4.1% and 1.5% in the SEMS and PS groups, respectively (OR, 1.56; 95% CI, 0.53–4.62; p=0.42).

Fig. 3.

Adverse event of stent placement for preoperative biliary drainage. (A) Endoscopic image of cholangitis caused by occlusion of a plastic stent. (B) A computed tomography (CT) scan image showing acute pancreatitis caused by placement of a self-expandable metal stent (SEMS). Peripancreatic effusion associated with post-ERCP pancreatitis (arrowheads). (C) A CT scan image showing acute cholecystitis caused by SEMS placement. Gallbladder wall thickening with pericholecystic inflammatory spread (arrowheads).

SEMSS VS. PSS IN TERMS OF POSTOPERATIVE ADVERSE EVENTS

In PBD for patients with pancreatic cancer receiving NAC, the incidence of postoperative adverse events was nearly equivalent between MS and PS (Table 1). A meta-analysis evaluated the safety of stents for preoperative drainage in patients with pancreatic cancer by comparing patients with PS placement and those with SEMS placement in terms of blood loss, operative time, postoperative adverse events, and surgical mortality.34,35 Blood loss did not significantly differ between the SEMS and PS groups (mean difference [MD], 43.0 mL; 95% CI, –202.1 to 288.2 mL; p=0.73). Similarly, operative time did not significantly differ between the SEMS and PS groups (MD, 18.0 minutes; 95% CI, –29.4 to 65.5 minutes; p=0.46). An analysis of 11 studies involving 571 patients found no significant difference in the incidence of postoperative adverse events between the SEMS and PS groups (OR, 0.78; 95% CI, 0.53–1.15; p=0.21). An analysis of six studies involving 315 patients found no significant difference in surgical mortality between the SEMS and PS groups (OR, 0.59; 95% CI, 0.10–3.60; p=0.57).35 These meta-analyses did not reveal a significant difference in surgical safety between SEMSs and PSs for PBD. However, this was not limited to patients with pancreatic cancer who underwent NAC. Furthermore, most studies focusing on patients with pancreatic cancer who received NAC had small sample sizes, indicating that larger studies are required to evaluate the safety of surgery.

MEDICAL COSTS

The lower cost of PSs compared with SEMSs is considered an advantage of PSs. However, in patients undergoing NAC, a prospective randomized controlled trial reported no significant difference in the total medical costs associated with biliary drainage between the stent types. Furthermore, no cost differences were observed when the entire perioperative period from surgery to hospital discharge was included.

HOW TO PREVENT ADVERSE EVENTS ASSOCIATED WITH PLACEMENT OF SEMSS

SEM placement for PBD in patients with pancreatic cancer is expected to reduce the incidence of time to RBO compared with PS placement. However, it is associated with an increased incidence of adverse events such as pancreatitis and cholecystitis following stent placement. Understanding the risk factors for post-ERCP pancreatitis and cholecystitis after SEMS placement is essential. A large multicenter study demonstrated that a pancreatic duct without tumor involvement (OR, 2.07; 95% CI, 1.30–3.31; p=0.023), an intact papilla (OR, 1.65; 95% CI, 1.06–2.44; p=0.025), and SEMS placement across the papilla (OR, 8.63; 95% CI, 1.14–65.07; p=0.037) were independent risk factors for acute pancreatitis, while tumor involvement at the orifice of the cystic duct (OR, 3.17; 95% CI, 1.80–5.54; p<0.001) was an independent risk factor for cholecystitis.38 In other studies, factors such as a narrow common bile duct diameter, the presence of gallbladder stones, and tumor invasion into the feeding artery of the gallbladder were reported as independent risk factors for cholecystitis after SEMS placement.39,40 In addition, Nakai et al.41 reported that axial force and SEMS length were associated with the incidence of cholecystitis, with high axial force and SEMS length ≤60 mm identified as independent risk factors for cholecystitis.

Additionally, there are methods to reduce the incidence of post-ERCP pancreatitis and acute cholecystitis after placement of SEMSs.42,43 It has been reported that placement of a pancreatic duct stent during deployment of a CSEMS for MBO significantly reduces the incidence of post-ERCP pancreatitis.43 Placement of a pancreatic duct stent is an effective method to prevent post-ERCP pancreatitis during CSEMS placement. However, in patients with pancreatic duct obstruction, such as those with pancreatic cancer, placement of a pancreatic duct stent can sometimes be difficult. Similarly, the effectiveness of stent placement in the cystic duct for preventing cholecystitis associated with CSEMS placement has been reported.42 The technical success rate of stent placement in the cystic duct in patients with distal MBO is 75%.42 Therefore, similar to pancreatic duct stent placement, cystic duct stenting is an effective method for preventing cholecystitis but can be technically difficult. For PBD, it has been reported the use of a CSEMS with a diameter of 6 mm, instead of the more commonly used 10-mm SEMS for MBO drainage, can reduce the incidence of pancreatitis and cholecystitis associated with SEMS placement.38,44,45 However, the incidence of RBO using CSEMSs with a 6-mm diameter for PBD in patients with pancreatic cancer receiving NAC tended to be higher (33.3%–45.3%) than the reported incidence of RBO using SEMSs with a 10-mm diameter.35,44,45

WHICH TYPE OF SEMS IS SUITABLE FOR PBD?

SEMSs have two main types. One is a CSEMS, which is covered with polyurethane, and the other is UCSEMS, which lacks this covering.38 Kitano et al.46 showed that the time to RBO was longer with CSEMS placement than with UCSEMS placement for ERBD in patients with distal MBO caused by pancreatic cancer. In addition, a meta-analysis showed that CSEMS is superior to UCSEMS in preventing RBO.47 In contrast, Seo et al.48 reported that in patients with distal MBO caused by pancreatic cancer who received NAC, the rate of RBO during PBD did not differ between CSEMSs and UCSEMSs. RBO was caused by tumor ingrowth in 16.7% of patients with UCSEMSs and by stent migration in 6.8% of patients with FCSEMSs.48 CSEMSs offer an advantage over UCSEMSs in terms of removability. One study indicated that CSEMSs are superior to UCSEMSs for PBD in patients with pancreatic cancer receiving NAC.29 Few studies have compared CSEMSs and UCSEMSs for PBD in patients with pancreatic cancer who underwent NAC. Therefore, whether CSEMSs or UCSEMSs are more suitable for preoperative drainage in this patient population remains controversial.

CONCLUSIONS

During PBD in patients with pancreatic cancer receiving NAC, maintaining drainage without the occurrence of RBO or cholangitis is crucial. The use of SEMSs for PBD significantly reduces the incidence of RBO and cholangitis compared with the use of PSs. A few reports have indicated that the rate of postoperative adverse events is higher in patients who underwent SEMS placement for PBD. Therefore, SEMS placement is more suitable for PBD in patients with pancreatic cancer receiving NAC. However, adverse events other than ERBD-related cholangitis, such as post-ERCP pancreatitis and cholecystitis, appear to occur more frequently with SEMS than with PS placement. Recognizing these risks and taking measures to prevent adverse events when using SEMSs for PBD are essential.

Article information Continued