11.5-Fr Large-Bore Plastic Stents Versus Fully Covered Self-Expandable Metallic Stents for the Management of Preoperative Endoscopic Biliary Obstruction in Pancreatic Cancer: A Multicenter Retrospective Cohort Study

Abstract

Background

Self-expandable metallic stents (SEMSs) are recommended for preoperative biliary drainage in patients with pancreatic cancer. However, they are associated with a relatively high incidence of non-recurrent biliary obstruction (RBO) adverse events (AEs), such as pancreatitis and cholecystitis. The 11.5-Fr large-bore plastic stent (LBPS) is a newly developed stent with a larger diameter than conventional plastic stents. The aim of the present study was to compare the clinical outcomes of LBPS and SEMS in preoperative biliary drainage for pancreatic cancer.

Methods

We retrospectively evaluated 80 patients who underwent preoperative biliary drainage for pancreatic cancer between January 2011 and December 2025. Patients were divided into the LBPS and SEMS groups according to the stent placed.

Results

RBO occurred in 20.0% and 25.0% of patients in the LBPS and SEMS groups, respectively, with no significant difference in the time to RBO between the groups. Pancreatitis and cholecystitis were observed in 0% versus 18.3%, and 0% versus 8.3% of patients in the LBPS and SEMS groups, respectively.

Conclusion

The 11.5-Fr LBPS demonstrated patency comparable to that of SEMS and was associated with a low incidence of non-RBO AEs. These findings should be interpreted cautiously and require confirmation in larger prospective studies.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s10620-026-10136-z.

Introduction

Pancreatic cancer often causes obstructive jaundice, for which endoscopic biliary drainage using endoscopic retrograde cholangiopancreatography (ERCP) is the treatment of choice. Plastic stents (PS) have traditionally been used for preoperative biliary drainage in patients with pancreatic cancer. The advantages of PS include their relative ease of placement, ability to be replaced and removed, and low cost. In recent years, however, preoperative neoadjuvant chemotherapy (NAC) has become the standard treatment, resulting in a longer preoperative waiting period compared with conventional upfront surgery [1–5]. Given this extended waiting period associated with NAC, the patency of PS is often insufficient, and the European Society of Gastrointestinal Endoscopy (ESGE) guidelines therefore recommend a 10-mm self-expandable metallic stent (SEMS) for preoperative drainage [6]. Although SEMS has a large caliber and is expected to prolong patency, they have been reported to be associated with a high incidence of non-recurrent biliary obstruction (RBO)-related adverse events (AEs), such as pancreatitis and cholecystitis [7–9]. These AEs raise concerns regarding potential discontinuation of NAC or postponement of surgery. Furthermore, evidence regarding the effects of SEMS on surgical outcomes remains limited. In contrast, the 11.5-Fr large-bore plastic stent (LBPS) (Oasis® One Action Stent Introduction System Preloaded with Cotton-Leung® Biliary Stent, Cook Medical Japan G.K.) is a newly developed stent with a larger diameter than conventional PS and three drainage lumens at the tip. Notably, the side holes are larger than those of conventional stents, suggesting improved drainage efficacy (Fig. 1a.b). The purpose of this study was to compare the clinical outcomes of LBPS and SEMS in preoperative biliary drainage for pancreatic cancer.

Fig. 1.

a, b 11.5-Fr LBPS (Oasis® One Action Stent Introduction System Preloaded with Cotton-Leung® Biliary Stent, Cook Medical Japan G.K.). There are three drainage lumens at the tip of the stent (red arrows). LBPS large-bore plastic stent

Patients and Methods

Patients’ Selection

This retrospective cohort study was conducted at four tertiary care centers: Gifu University Hospital, Gifu Prefectural General Medical Center, Gifu Municipal Hospital, and Matsunami General Hospital. A database, which encompassing all patients who were histologically diagnosed with pancreatic cancer between January 2011 and December 2025, was analyzed to identify those who met the following inclusion criteria: [1] pathologically confirmed pancreatic cancer classified as resectable or borderline resectable based on computed tomography (CT), magnetic resonance imaging (MRI), or endoscopic ultrasound-guided fine needle aspiration or biopsy (EUS-FNA/FNB); [2] underwent preoperative biliary drainage for obstructive jaundice due to pancreatic cancer using either LBPS or 8–12 mm fully covered self-expandable metallic stent (FCSEMS); and [3] age older than 20 years. Patients were excluded if they met any of the following exclusion criteria: (1) patients who underwent palliative biliary drainage in non-resectable cases, (2) biliary drainage with PS smaller than 11.5-Fr LBPS, (3) biliary drainage performed using percutaneous transhepatic biliary drainage (PTBD) or endoscopic ultrasound-guided biliary drainage (EUS-BD), except for the rendezvous technique; (4) coexisting malignant biliary obstruction involving the hilar part of the biliary system, defined as within 2 cm below the biliary confluence; (5) biliary drainage performed at other institutions; (6) a prior history of surgical biliary reconstruction; or (7) adverse events (AEs), such as hemorrhage or perforation, occurring before stent deployment. Patients deemed ineligible for the study by the principal investigator and those who declined participation based on publicly available information were also excluded. Pancreatic cancer staging was performed according to the Union for International Cancer Control (UICC) criteria [10]. Resectability was determined based on the clinical practice guidelines for pancreatic cancer issued by the Japan Pancreas Society [11, 12]. Patients were divided into the LBPS and SEMS groups according to the type of stent used. Written informed consent was obtained from all patients prior to the procedures, and consent for the use of research data was obtained on an opt-out basis. All procedures in this study were conducted in accordance with the ethical standards of the institutional and national research committees and the principles of the 1964 Helsinki Declaration and its later amendments. The study protocol was approved by the Institutional Review Board of each participating institution (Gifu University Hospital: 2025–359) and was registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN: 60443).

Endoscopic Procedures

All ERCP procedures were performed using a standard duodenoscope (TJF-260 V, JF-260 V, or JF-240; Olympus, Tokyo, Japan) under conscious sedation with midazolam and pentazocine, with continuous monitoring of vital signs. Bile duct cannulation was performed using a wire-loaded cannulation technique. If biliary cannulation was difficult with the standard approach, the pancreatic guidewire technique or pre-cutting was applied as needed. After successful selective bile duct cannulation, cholangiography was performed to evaluate the location and length of the malignant biliary obstruction. Endoscopic sphincterotomy (EST) or two-step drainage using endoscopic nasobiliary drainage (ENBD) was performed at the operator’s discretion. EST was performed using a papillotome (KD-211Q-0725 or KD-V411M-0725; Olympus, Tokyo, Japan, or Correctome; Boston Scientific, Massachusetts, USA) over a guidewire. A stent of sufficient length to cover both the obstruction and the duodenal papilla was selected for biliary drainage. Subsequently, a FCSEMS (8–12 mm in diameter and 5–8 cm in length) or LBPS (11.5-Fr and 5–8 cm in length) was deployed at the obstruction site. The diameter, length, and type of stent were chosen at the discretion of each endoscopist. Antibiotics, nonsteroidal anti-inflammatory drugs (NSAIDs), and protease inhibitors were administered before or after ERCP according to institutional protocols.

Neoadjuvant Chemotherapy

After proper biliary drainage and improvement of liver function, NAC was performed in some patients. The regimens used were S-1, gemcitabine plus S-1, gemcitabine plus nab-paclitaxel, or FOLFIRINOX according to the discretion of each hospital. Contrast-enhanced CT or MRI is generally performed for restaging after the completion of NAC. The final decision to perform surgical exploration was made by an expert panel of endoscopists, surgeons, and radiologists.

Surgical Resection

The standard surgical procedures included pylorus-preserving pancreaticoduodenectomy (PPPD) or subtotal stomach-preserving pancreaticoduodenectomy (SSPPD) with standard lymphadenectomy. In addition, classic Child reconstruction was performed with resection of the distal stomach. If peritoneal metastasis or major vascular invasion was identified during surgery, palliative treatment consisting of gastrojejunostomy was performed.

Study Outcomes, Definitions, and Statistical Analyses

The primary outcomes of this study were the rate of RBO and the time to RBO (TRBO) until surgery in the LBPS and SEMS groups. Secondary outcomes included the technical success rate, clinical success rate, ERCP- and stent-related AEs, and surgery-related outcomes in the two groups. Technical success was defined as successful biliary stent placement at the appropriate location. Clinical success was defined as a ≥ 50% reduction or normalization of total bilirubin or liver enzyme levels within 14 days. The definitions and severity grading of ERCP- or stent-related AEs were evaluated according to the 2024 Tokyo Criteria for Transpapillary Biliary Stenting [13]. RBO was defined as stent occlusion, stent migration, or any condition requiring re-intervention, such as biliary drainage or stent removal. Stent migration was defined as stent dislocation confirmed by endoscopic findings or imaging studies. Stent occlusion included tumor overgrowth, tumor ingrowth, sludge formation, and food impaction. TRBO was defined as the time interval from stent placement to the occurrence of RBO. Non-RBO AEs, including pancreatitis, cholangitis, cholecystitis, bleeding, and perforation, were determined based on the 2024 TOKYO criteria. Non-RBO AEs were defined as events requiring conservative treatment, medication, intervention, or hospitalization.

Surgery-related outcomes included the curative resection rate, operative time, intraoperative blood loss, and the incidence of postoperative AEs. Postoperative AEs were graded according to the Clavien–Dindo (CD) classification and the definitions of the International Study Group on Pancreatic Fistula; AEs greater than Grade III were considered significant [14, 15]. A clinically curative resection was defined as R0 surgery, in which no cancer cells were observed microscopically at the surgical margins.

Continuous variables were reported as medians with ranges or interquartile ranges (IQRs). Categorical variables were compared using Fisher’s exact test or Pearson’s chi-squared test, as appropriate, and continuous variables were compared using the Mann–Whitney U test. TRBO was estimated using the Kaplan–Meier method, and cumulative incidence was compared between the LBPS and SEMS groups using the log-rank test. A Cox proportional hazards model was used to estimate hazard ratios (HRs) for RBO, with 95% confidence interval (CIs). Factors with a P < 0.20 in univariable analysis and stent type were further assessed using multivariable analysis. Statistical significance was set at P < 0.05. All statistical analyses were performed using JMP version 18 (SAS Institute 2, Inc., Cary, NC, USA).

Results

Patient Selection

A total of 238 patients underwent preoperative drainage for obstructive jaundice caused by pancreatic cancer at the 4 centers during the study period. Among these patients, 158 were excluded for the following reasons: biliary drainage with PS smaller than 11.5-Fr in 144 patients, initial biliary drainage (PTBD or EUS-BD) in five patients, biliary drainage at other centers in four patients, AEs before stent deployment in three patients, and refusal of treatment due to old age in two patients. After screening, 80 patients fulfilled the eligibility criteria and were included in this study; for biliary drainage, LBPS was placed in 20 patients and SEMS in 60 patients. (Fig. 2).

Fig. 2.

Flowchart of the patient selection process. AEs adverse events, BSC best supportive care, EUS-BD endoscopic ultrasound-guided biliary drainage, LBPS large-bore plastic stent, PTBD percutaneous transhepatic biliary drainage, SEMS self-expandable metallic stent

Baseline Characteristics and Endoscopic Procedures

Patient characteristics of the two groups are summarized in Table 1. Significant differences between the two groups were identified with respect to the initial total albumin level, history of endoscopic sphincterotomy (EST), pre-drainage with endoscopic nasobiliary drainage (ENBD), and biliary stricture length. The endoscopic procedures performed in the two groups are summarized in Table 2. No significant differences were observed between the groups with respect to the cannulation method, pancreatography, pancreatic stent, and EST. There were more prophylactic rectal NSAIDs administered in the LBPS group. In the latter half of this study period, rectal NSAIDs were actively used, with a 95% usage rate in the LBPS group. On the other hand, in the SEMS group, there were cases before the introduction of rectal NSAIDs, and immediately after introduction, the decision to use them was judged by the operator, such as using them only in cases with a high risk of pancreatitis.

Table 1.

Patient characteristics of the two groups

	LBPS group (N = 20)	SEMS group (N = 60)	P value
Age, y.o, median (IQR)	69 (57–83)	70 (63–77)	0.643
Sex, Male, n (%)	11 (55.0)	30 (50.0)	0.798
Initial total bilirubin level, mg/dl, median (IQR)	4.6 (1.6–8.6)	6.1 (1.7–10.1)	0.961
Initial total albumin level, mg/dl, median (IQR)	3.8 (3.5–3.9)	4.0 (3.8–4.3)	0.007
CA19-9, U/ml, median (IQR)	251 (93–913)	310 (64–796)	0.322
Primary tumor diameter, mm, median (IQR)	21 (18–25)	22 (18–25)	0.907
Diameter of main pancreatic duct, mm, median (IQR)	4 (3–6)	4 (3–7)	0.776
History of pancreatitis, n (%)	0	0	> 0.99
History of EST, n (%)	0	26 (43.3)	< 0.001
Pre-drainage with ENBD, n (%)	0	30 (50.0)	< 0.001
Pancreatic atrophy, n (%)	3 (15.0)	22 (36.6)	0.070
Dilated bile duct diameter, mm, median (IQR)	14 (11–16)	12 (11–15)	0.134
Biliary stricture length, mm, median (IQR)	18 (12–33)	14 (10–20)	0.031
Primary clinical stage, n (%)			0.743
IA	9 (45.0)	25 (41.6)
IB	8 (40.0)	21 (35.0)
IIA	0	3 (5.0)
IIB	3 (15.0)	11 (18.3)
Resectability, n (%)			> 0.99
R	14 (70.0)	40 (66.6)
BR	6 (30.0)	20 (33.3)
Neoadjuvant chemotherapy, n (%)	11 (55.0)	31 (51.6)	> 0.99
GS	7	22
GnP	4	5
mFX	0	3
S-1	0	1
Radiation therapy, n (%)	0	9 (15.0)	0.102

BR borderline resectable, ENBD endoscopic nasobiliary drainage, GnP gemcitabine plus nab-paclitaxel, GS gemcitabine plus S-1, IQR interquartile range, LBPS large-bore plastic stent, mFX modified FOLFIRINOX, R resectable, SEMS self-expandable metallic stent

Table 2.

ERCP-related outcomes of the two groups

	LBPS group (N = 20)	SEMS group (N = 60)	P value
Prophylactic rectal NSAIDs, n (%)	19 (95.0)	28 (46.6)	< 0.001
Biliary cannulation method, n (%)			0.645
Standard	10 (50.0)	28 (46.6)
Guidewire-assisted	8 (40.0)	18 (30.0)
Pancreatic guidewire-assisted	1 (5.0)	7 (11.6)
Precut	1 (5.0)	3 (5.0)
EUS-RV	0	4 (6.6)
Pancreatography, n (%)	5 (25.0)	20 (33.3)	0.584
Pancreatic duct stent, n (%)	0	5 (8.3)	0.323
EST, n (%)	20 (100)	53 (88.3)	0.183
Stent size, n (%)			< .0001
11.5Fr	20 (100)	0
8 mm	0	5 (8.3)
10 mm	0	37 (61.6)
12 mm	0	18 (30.0)
Stent length, n (%)			< .0001
5 cm	4 (20.0)	6 (10.0)
6 cm	0	35 (58.3)
7 cm	16 (80.0)	8 (13.3)
8 cm	0	11 (18.3)
Technical success, n (%)	20 (100)	60 (100)	> 0.99
Clinical success, n (%)	19 (95.0)	56 (93.3)	> 0.99
AEs, n (%)	6 (30.0)	26 (43.3)	0.429
RBO, n (%)	4 (20.0)	15 (25.0)	0.768
Non-RBO AEs, n (%)	2 (10.0)	17 (28.3)	0.132
Acute pancreatitis, n (%)	0	11 (18.3)	0.057
Acute cholecystitis, n (%)	0	5 (8.3)	0.323
Non-obstructive cholangitis, n (%)	2 (10.5)	2 (3.3)	0.258

AEs adverse events, ERCP endoscopic retrograde cholangiopancreatography, EST endoscopic sphincterotomy, EUS-RV endoscopic ultrasound-guided rendezvous, LBPS large-bore plastic stent, NSAIDs nonsteroidal anti-inflammatory drugs, RBO recurrent biliary obstruction, SEMS self-expandable metallic stent

ERCP-Related Outcomes

The ERCP-related outcomes of the two groups are summarized in Table 2. The technical success rate was 100% in both groups. The clinical success rates were 95.0% (19/20) in the LBPS group and 93.3% (56/60) in the SEMS group, respectively (P > 0.99). In the five cases of clinical failure (one in the LBPS exchange group and four in the SEMS group), all patients experienced early stent occlusion or migration within two weeks of placement, all patients underwent stent exchange. The overall incidence of AE related to ERCP and stent placement was 30.0% in the LBPS group and 43.3% in the SEMS group, respectively (P = 0.429). RBO occurred in 20.0% and 25.0% of the patients in the LBPS and SEMS groups, respectively (P = 0.768). In the Kaplan–Meier analysis, there was no significant difference in the time to RBO (TRBO) between the two groups (hazard ratio for RBO, 0.567; 95% CI, 0.186–1.727; log-rank P = 0.310; Fig. 3). In univariable and multivariable analyses, only a dilated bile duct was identified as a risk factor for RBO; however, LBPS itself was not a risk factor for RBO (Table 3). Other than RBO, pancreatitis, cholecystitis, and non-obstructive cholangitis were observed in 0% vs. 18.3% (P = 0.057), 0% vs. 8.3% (P = 0.323), and 10.5% vs. 3.3% (P = 0.258), respectively. In the SMES group, there were nine mild cases and two moderate cases of pancreatitis, which improved with conservative treatment without developing local complications such as walled-off necrosis. Of the five cases of cholecystitis, two were successfully managed with conservative treatment, but three required percutaneous transhepatic gallbladder drainage (PTGBD). Two cases of non-obstructive cholangitis in the LBPS group and two cases in the SEMS group were successfully treated with antibiotics. In univariable and multivariable analyses, LBPS placement showed an association with a lower incidence of post-ERCP pancreatitis (Table 4), although the limited number of events precludes firm conclusions regarding independent risk factors.

Fig. 3.

Kaplan–Meier analysis of the incidence rate of RBO. LBPS large-bore plastic stent, RBO recurrent biliary obstruction, SEMS self-expandable metallic stent, 95% CI 95% confidence interval

Table 3.

Univariable and multivariable analysis for risk factors associated with RBO

				Univariable analysis		Multivariable analysis
		No of cases	No of events	OR (95% CI)	p value	OR (95% CI)	p value
Age	≥ 70	42	12 (28.5)	0.564 (0.187–1.598)	0.284	–	–
Sex	Male	41	8 (19.5)	1.620 (0.576–4.724)	0.360	–	–
Initial total bilirubin level, mg/dl	≥ 5.5	40	10 (25.0)	0.870 (0.305–2.453)	0.792	–	–
Initial total albumin level, mg/dl	≥ 3.9	45	11 (24.4)	0.915 (0.314–2.582)	0.868	–	–
CA19-9, U/ml	≥ 297	40	11 (27.5)	0.659 (0.226–1.852)	0.429	–	–
Primary tumor diameter, mm	≥ 22	40	8 (20.0)	1.517 (0.539–4.419)	0.429	–	–
Dilated bile duct diameter, mm	≥ 13	40	12 (30.0)	0.198 (0.051–0.618)	0.004	0.198 (0.049–0.658)	0.007
Biliary stricture length, mm	≥ 15	41	15 (36.5)	0.494 (0.164–1.401)	0.186	0.415 (0.120–1.309)	0.135
Primary clinical stage	≥ II	17	7 (41.1)	0.336 (0.105–1.088)	0.068	0.355 (0.098–1.267)	0.109
Neoadjuvant chemotherapy	Yes	42	11 (26.1)	0.751 (0.258–2.113)	0.589	–	–
EST	Yes	73	18 (24.6)	0.509 (0.026–3.264)	0.516	–	–
Stent type	LBPS	20	4 (20.0)	1.333 (0.410–5.201)	0.644	2.000 (0.538–8.894)	0.310

EST endoscopic sphincterotomy, LBPS large-bore plastic stent, OR odds ratio, 95% CI 95% confidence interval

Table 4.

Univariable and multivariable analysis for risk factors associated with post-ERCP pancreatitis

				Univariable analysis		Multivariable analysis
		No of cases	No of events	OR (95% CI)	p value	OR (95% CI)	p value
Age	≥ 70	42	6 (14.2)	1.1 (0.303–4.139)	0.883	–	–
Sex	Male	41	8 (19.5)	2.909 (0.769–14.13)	0.118	2.275 (0.513–12.23)	0.284
Initial total albumin level, mg/dl	≥ 3.9	45	10 (22.2)	0.102 (0.005–0.581)	0.006	0.201 (0.010–1.275)	0.094
Primary tumor diameter, mm	≥ 22	40	4 (10.0)	1.909 (0.527–7.842)	0.327	–	–
Pancreatic atrophy	Yes	25	1 (4.0)	0.187 (0.009–1.068)	0.061	0.204 (0.010–1.364)	0.107
Prophylactic rectal NSAIDs	Yes	47	6 (12.7)	0.819 (0.225–3.090)	0.761	–	–
Biliary cannulation method	Pancreatic guidewire-assisted / Precut	12	3 (25.0)	2.5 (0.482–10.63)	0.252	–	–
Pancreatography	Yes	25	4 (16.0)	1.306 (0.314–4.817)	0.696	–	–
Pancreatic duct stent	Yes	5	1 (20.0)	1.625 (0.078–12.46)	0.690
Stent type	LBPS	20	0	NA	0.008	NA	0.028

LBPS large-bore plastic stent, NA not applicable, NSAIDs nonsteroidal anti-inflammatory drugs, OR odds ratio, 95% CI 95% confidence interval

Surgery-Related Outcomes

The median time to surgery was 64 days (IQR, 45–78 days) in the LBPS group and 73 days (IQR, 29–106 days) in the SEMS group, with no significant differences (P = 0.426). The surgery-related outcomes of the two groups are summarized in Table 5. Surgical resection was successful in 60.0% in the LBPS group and 80.0% in the SEMS group, respectively (P = 0.133). In the LBPS group, among the 8 cases in which curative surgery was not performed, 4 cases experienced tumor progression such as local progression or distant metastasis during NAC, 1 case had positive intraoperative ascites cytology, 1 case was prioritized for treatment of other diseases, and 2 cases did not wish to undergo surgery. Regarding surgical procedures, SSPPD was significantly more common in the LBPS group; however, the curative surgery rates were 91.6% and 72.9% in the LBPS and SEMS groups, respectively, with no significant differences. The median surgery times were 409 and 542 min in the LBPS and SEMS groups, respectively, and were significantly shorter in the LBPS group (P < 0.001). Surgery-related AE occurred in 16.6% (2/12: pancreatic fistula in two patients) of the LBPS group and 14.5% (7/48: pancreatic fistula in one patient, biloma in three patients, abscess in one patient, and others in four patients) of the SEMS group (P > 0.99). The median time of postoperative hospital stay in the LBPS and SEMS groups was 21 and 27 days, respectively, and tended to be shorter in the LBPS group (P = 0.080).

Table 5.

Surgery-related outcomes of the two groups

	LBPS group (N = 12)	SEMS group (N = 48)	P value
Time to surgery, days, median (IQR)	64 (45–78)	73 (29–106)	0.426
Resection success, n (%)	12 (60.0)	48 (80.0)	0.133
Surgical procedure, n (%)			0.030
SSPPD	12/12 (100)	29/48 (60.4)
PPPD	0	18/48 (37.5)
PD	0	1/48 (2.0)
Curative surgery, R0, n (%)	11/12 (91.6)	35/48 (72.9)	0.261
Intraoperative bleeding, ml, median (IQR)	721 (582–800)	891 (503–1528)	0.149
Surgery time, min, median (IQR)	409 (391–424)	542 (502–624)	< 0.001
Surgery-related AEs, n (%)	2/12 (16.6)	7/48 (14.5)	> 0.99
Pancreatic fistula	2	1
Biloma	0	3
Abscess	0	1
Others	0	4
Surgery-related death, n (%)	0	0	> 0.99
Histology, n (%)			0.229
Well differentiated	6/12 (50.0)	12 (25.0)
Moderately differentiated	5 (41.6)	28 (58.3)
Poorly differentiated	0	6 (12.5)
Others	1/12 (8.3)	2 (4.1)
Final stage, n (%)			0.117
IA	5/12 (41.6)	9/48 (18.7)
IB	4/12 (33.3)	7/48 (14.5)
IIA	0	1/48 (2.0)
IIB	2/12 (16.6)	27/48 (56.2)
III	1/12 (8.3)	4/48 (8.3)
Postoperative hospital stay, days, median (IQR)	21 (15–27)	27 (23–37)	0.080

AEs adverse events, IQR interquartile range, LBPS large-bore plastic stent, PD pancreaticoduodenectomy, PPPD pylorus-preserving pancreaticoduodenectomy, RBO recurrent biliary obstruction, SEMS self-expandable metallic stent, SSPPD subtotal stomach-preserving pancreaticoduodenectomy

Discussion

Summary of the Results of This Study

This study compared the clinical outcomes of LBPS and SEMS placement for the preoperative management of biliary obstruction in patients with resectable pancreatic cancer. RBO occurred in 20.0% and 25.0% of the patients in the LBPS and SEMS groups, respectively, which was not significantly different. TRBO was similar in both groups, and in a multivariable analysis examining factors affecting RBO, the type of stent itself was not identified as a significant factor. In contrast, the incidence rates of pancreatitis and cholecystitis were tended to be higher in the SEMS group. Regarding surgery-related outcomes, no significant differences were observed between the two groups in curative surgery rate, intraoperative bleeding volume, and AEs, but the surgical time was significantly shorter in the LBPS group, and the length of hospital stay also tended to be shorter in the LBPS group.

Usefulness and Issues of SEMS in Preoperative Biliary Drainage for Pancreatic Cancer

In recent years, NAC has become the mainstream preoperative treatment for pancreatic cancer, and the waiting period before surgery has been extended [1–5]. Before the introduction of NAC, the waiting period for surgery was between 12 and 36 days; however, it has been reported that this extended to 72–165 days after the introduction of NAC [16]. PS has traditionally been used as a biliary stent for preoperative drainage in patients with pancreatic cancer; however, SEMSs are now recommended considering the extended preoperative period. Kobayashi et al. retrospectively compared the SEMS group (n = 21) and PS group (n = 22) for preoperative biliary drainage caused by pancreatic cancer [17]. In the SEMS group, 8-mm fully covered metallic stents were used, whereas in the PS group, 7-Fr plastic stents were used. During the waiting period for surgery, RBO occurred in 4.8% and 95.4% of patients in the SEMS and PS groups, respectively (p < 0.001). The TRBO was significantly longer in the SEMS group than in the PS group. These results suggest that SEMSs have a significantly longer patency than PS and are useful as biliary drainage stents for preoperative pancreatic cancer. However, a disadvantage of SEMS is the high incidence of AEs other than RBO, such as pancreatitis and cholecystitis [18–21]. A retrospective study by Ichikawa et al. compared SEMS (n = 45) and PS (n = 75) groups for preoperative biliary drainage in patients with pancreatic cancer [18]. The RBO rate was significantly lower in the SEMS group (24.4% vs. 44.0%, P = 0.034), and the TRBO was significantly longer in the SEMS group. However, pancreatitis and cholecystitis were observed in 15.6% and 2.7% (P = 0.026) and 6.7% and 0% (P = 0.051) in the SEMS and PS groups, respectively. Cote et al. conducted a retrospective study to investigate the risk factors of post-ERCP pancreatitis (PEP) [20]. They enrolled 544 patients who underwent stent placement for malignant biliary obstruction, with 248 and 296 patients in the SEMS and PS groups, respectively. The frequency of PEP was significantly higher in the SEMS group (7.3%) than in the PS group (1.3%) (P < 0.01). The results showed that the frequency of post-ERCP pancreatitis was significantly higher in the SEMS group. Cao et al. retrospectively examined the risk factors of post-ERCP cholecystitis in 4238 patients who underwent their first ERCP [21]. Univariate and multivariate analyses identified biliary duct SEMS placement as a risk factor for post-ERCP cholecystitis (OR, 3.66; 95% CI 1.78–7.54). In the current study, pancreatitis and cholecystitis were observed in 0% vs. 18.3% and 0% vs. 8.3% of the patients in the LBPS and SEMS groups, respectively, and tended to be higher in the SEMS group. These results suggest that SEMS for preoperative biliary drainage of pancreatic cancer has a significantly higher incidence of AEs, such as pancreatitis and cholecystitis, than PS, and there is concern that these AEs may lead to the postponement of NAC or surgery itself. In other words, the relationship between stent patency based on caliber and AEs other than RBO (pancreatitis and cholecystitis) is thought to be a trade-off. In preoperative drainage for pancreatic cancer, a stent strategy that takes into consideration not only RBO but also AEs such as pancreatitis and cholecystitis is required.

Outcomes of Preoperative Biliary Drainage Using Large-Bore PS for Pancreatic Cancer and Its Impact on Surgery

The PS used for preoperative biliary drainage of pancreatic cancer has insufficient patency, and to compensate for this weakness, it is necessary to increase its caliber. Traditionally, a 7-Fr PS has been used primarily for the preoperative drainage of pancreatic cancer, but there have been several reports examining the results of upgrading to a 10-Fr PS. Mandai et al. conducted a multicenter randomized controlled trial (RCT) to compare the clinical outcomes of 10-mm SEMS (n = 17) and 10-Fr PS (n = 22) for preoperative biliary drainage in pancreatic cancer [19]. The re-intervention rates were 0% in the SEMS group and 29.4% in the PS group, and patency was significantly longer in the SEMS group than in the PS group. Another RCT by Tamura et al. evaluated a 10-mm SEMS (n = 11) and 10-Fr PS (n = 11) for preoperative biliary drainage in ancreatic cancer [22]. The SEMS group showed a significantly lower rate of RBO (18.2% vs. 72.8%, P = 0.015), longer stent patency, and fewer reinterventions for stent dysfunction than the PS group. These results suggest that the caliber of a 10-Fr PS is insufficient for preoperative biliary drainage in patients with pancreatic cancer, and that its patency is significantly shorter than that of the SEMS. However, it should be noted that all these reports had relatively small sample sizes.

The LBPS used in this study is a new type of plastic stent with a larger diameter of 11.5 Fr. The LBPS has three drainage lumens, and the side holes are larger than those of the conventional PS; therefore, excellent drainage effects can be expected. In the current study, the incidence of RBO during the waiting period for surgery in the LBPS group was comparable to that in the SEMS group, and no significant difference in TRBO was observed between the two groups. Furthermore, the incidence of AEs other than RBO, such as pancreatitis and cholecystitis, was 0% in the LBPS group, whereas the SEMS group had higher rates at 18.3% and 8.3%, respectively. These results suggest that LBPS does not have inferior RBO compared to SEMS and may reduce the incidence of adverse events other than RBO, such as pancreatitis and cholecystitis.　Although the incidences of pancreatitis and cholecystitis were numerically lower in the LBPS group, these differences did not reach statistical significance. Therefore, these findings should be interpreted with caution, as the present study may have been underpowered to detect clinically relevant differences.

However, there is currently insufficient evidence regarding the effects of PS and SEMS use on surgery. Tamura et al. conducted an RCT to compare the effects of SEMS (n = 7) and PS (n = 9) placement in pancreatic cancer surgery and found no significant differences between the two groups in terms of operation time, intraoperative blood loss, postoperative hospitalization, or postoperative AEs [22]. Another RCT by Mandai et al., which compared a SEMS group (n = 17) with a PS group (n = 22), revealed that the SEMS group had significantly more intraoperative blood loss and AE than the PS group, and the postoperative hospital stay was significantly longer in the SEMS group [19]. In the current study, no significant differences were observed between the two groups in curative surgery rate, intraoperative bleeding volume, and AEs, but the surgical time was significantly shorter in the LBPS group, and the length of hospital stay also tended to be shorter in the LBPS group. These results suggest that PS, including LBPS, is unlikely to have a negative effect on surgery and may actually have less of an effect than SEMS.

6-mm SEMS for Preoperative Biliary Drainage of Pancreatic Cancer

In recent years, the 6-mm SEMS has attracted attention as a preoperative drainage for pancreatic cancer, considering the trade-off between patency and AEs depending on the stent diameter. Kataoka et al. conducted a retrospective comparative study of the clinical outcomes of a 6-mm SEMS group (n = 26) and a PS group (n = 25) for preoperative biliary drainage for pancreatic cancer [23]. The incidence of RBO was significantly lower in the 6-mm SEMS group (7.7%) than that in the PS group (40.0%), and TRBO was significantly longer in the 6-mm SEMS group. The groups did not significantly differ in terms of AEs other than RBO, such as pancreatitis (15.4% vs. 4.0%) and cholecystitis (7.7% vs. 4.0%). Sakai et al. conducted a retrospective comparative study on the outcomes of 6-mm (n = 27) and 10-mm SEMS (n = 26) in preoperative biliary drainage for pancreatic cancer [24]. The RBO rate was 26.9% in the 6-mm SEMS group, which was significantly higher than that in the 10-mm SEMS group (7.7%), and the rate of migration was particularly high at 26.9%. Although there was no significant difference between the two groups in terms of non-RBO AEs, the incidence of pancreatitis was 3.7% vs. 11.5%, and the incidence of cholecystitis was 0% vs. 7.7%, both of which were higher in the 10-mm SEMS group. These results suggest that a 6-mm SEMS can maintain longer patency than a PS and that it can also reduce the occurrence of non-RBO complications such as pancreatitis and cholecystitis compared to a 10-mm SEMS. However, this also suggests that the 6-mm SEMS has a higher RBO owing to migration. In the current study, the 11.5-Fr LBPS demonstrated its usefulness and safety as a stent with patency equivalent to that of a SEMS and suppressed non-RBO AEs such as pancreatitis and cholecystitis; however, there have been no previous reports comparing it with a 6-mm SEMS. In many cases of suspected malignant biliary obstruction due to pancreatic cancer, a PS is often initially placed, if it is replaced with a SEMS after the diagnosis is confirmed, an additional ERCP session is required. However, if an LBPS is placed as the initial stent, additional ERCP is not required to exchange the stent after the diagnosis is confirmed, and NAC treatment can be initiated early. Thus, it is considered that there is great significance in selecting LBPS as the initial stent for preoperative biliary drainage in pancreatic cancer patients. Based on the results of this study, an 11.5-Fr LBPS may be the first choice for preoperative biliary drainage for pancreatic cancer. However, considering that current guidelines recommend SEMS, one option is to actively use it in cases at high risk of post-ERCP pancreatitis or cholecystitis. The American Society for Gastrointestinal Endoscopy (ASGE) and European Society of Gastrointestinal Endoscopy (ESGE) guidelines state that factors that increase the risk of pancreatitis include young age, female gender, and a history of acute pancreatitis or PEP [25, 26]. In addition, the risk of developing cholecystitis due to SEMS placement has been reported to include tumor involvement in the orifice of the cystic duct and intragallbladder stones [27–30]. In these cases, 11.5-Fr LBPS is considered a good option to avoid the development of pancreatitis and cholecystitis due to SEMS placement. This study suggests that LBPS is a promising alternative for preoperative biliary drainage, although the observed advantages should be interpreted as associations identified in this retrospective study rather than evidence of causality. There is currently no established strategy for determining the optimal stent type and diameter for preoperative biliary drainage of pancreatic cancer, and further investigation in larger prospective comparative studies is needed.

Limitation

This study has several limitations. First, the sample size is small. Second, it was retrospective and conducted at four tertiary care centers, which could potentially have introduced selection bias and reduced its external validity. Third, LBPS was used more frequently during the later phases of the study period. Temporal changes in ERCP techniques, prophylactic strategies for post-ERCP pancreatitis, perioperative management, and neoadjuvant treatment protocols may have influenced the observed outcomes. Therefore, residual confounding related to treatment era cannot be completely excluded. Fourth, there are differences in the number of curative surgeries performed in the two groups, limiting the interpretation of surgical outcomes. Furthermore, the impact of LBPS and SEMS placement on surgery may not have been fully evaluated using the current items alone. Finally, the number of post-ERCP pancreatitis events was limited relative to the number of variables included in the multivariable analysis. Therefore, the model may have been underpowered and susceptible to overfitting, and the identified association between LBPS placement and a lower incidence of post-ERCP pancreatitis should be interpreted with caution. Validation in larger prospective studies is warranted.

Conclusion

In conclusion, 11.5-Fr LBPS demonstrated patency comparable to that of SEMS. Although pancreatitis and cholecystitis occurred less frequently in the LBPS group, these differences were not statistically significant. Larger prospective randomized controlled studies are warranted to clarify whether LBPS provides a clinical advantage over SEMS.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AE

Adverse event

ERCP

Endoscopic retrograde cholangiopancreatography

EST

Endoscopic sphincterotomy

EUS-BD

Endoscopic ultrasound-guided biliary drainage

EUS-FNA

Endoscopic ultrasound-guided fine needle aspiration

EUS-FNB

Endoscopic ultrasound-guided fine needle biopsy

HR

Hazard ratio

IQR

Interquartile range

LBPS

Large-bore plastic stent

NAC

Neoadjuvant chemotherapy

PEP

Post-ERCP pancreatitis

PTBD

Percutaneous transhepatic biliary drainage

RBO

Recurrent biliary obstruction

SEMS

Self-expandable metallic stent

TRBO

Time to recurrent biliary obstruction

UICC

Union for International Cancer Control

95% CI

95% Confidence interval

Author Contributions

AM: wrote the manuscript. AM, TK, KS, YO, SI, HI, NM, YI, MO, SU, KI, TM, TI, and MS managed the patients. AM took the correspondence. All authors read and approved the final manuscript.

Funding

Open Access funding provided by Gifu University.

Data Availability

No datasets were generated or analyzed during the current study.

Declarations

Conflict of interest

The authors declare no conflicts of interest for this article.