Abstract
Background/Aims
Pancreatic duct (PD) stent selection impacts migration and retention. We evaluated risk factors for migration versus inappropriate retention of PD stents at a high volume center using a standardized stent management protocol.
Methods
Patients undergoing endoscopic retrograde cholangiopancreatography (ERCP) with PD stenting were retrospectively categorized by stent status at 2 weeks post-ERCP. Spontaneous migration was defined as absence of a PD stent on follow-up imaging, while inappropriate retention was defined as persistence of a non-flanged PD stent within the duct. The primary endpoint was PD stent retention at 2 weeks. Secondary endpoints included predictors of inappropriate retention and the need for endoscopic stent removal.
Results
A total of 482 patients underwent ERCP for biliary (n=289), pancreatic (n=181), or dual (n=12) indications. Non-flanged PD stents for biliary vs. pancreatic indications were 74.4% vs. 31.5% (p<0.001). Among 425 patients with follow-up imaging, predictors of PD stent retention included internal flanges (odds ratio [OR], 6.39; 95% confidence interval [CI], 3.29–12.67), pancreatic indication (OR, 2.23; 95%CI, 1.22–4.07), and stent length >3cm (OR, 2.5; 95% CI, 1.23–5.07). Inappropriate retention of non-flanged stents occurred in 29 patients (12.5%), with only 5 (2.2%) requiring endoscopic removal due to mechanical obstruction.
Conclusions
Inappropriate retention of short, narrow caliber, non flanged PD stents is infrequent and rarely requires endoscopic removal.
Keywords: Endoscopic retrograde cholangiopancreatography, Outcomes, Stents
Graphical abstract
INTRODUCTION
Endoscopic placement of stents in the pancreatic duct (PD) is performed for a wide variety of indications, ranging from the prevention of pancreatitis after endoscopic retrograde cholangiopancreatography (ERCP) to the treatment of leaks and strictures.1-5 Despite their beneficial role, PD stent placement is also associated with risks of PD injury, manifested as de novo strictures and morphological changes.6-8 Some data suggest that stents made of softer material and without internal flanges—widely used for prophylaxis of post-ERCP pancreatitis—may not necessarily cause ductal changes.8 Nonetheless, current guidelines recommend cross-sectional imaging to document spontaneous migration of prophylactically placed PD stents.9-11
The optimal timing of imaging to evaluate stent migration remains a matter of debate. The European Society of Gastrointestinal Endoscopy recommends imaging studies within 5–10 days after ERCP,9 whereas the American Society for Gastrointestinal Endoscopy (ASGE) summary document does not provide a specific timeline for imaging but suggests removal at 2–4 weeks after ERCP.10 The methodology document from the ASGE supports imaging within 2 weeks but also notes wide variation in published studies.11 Nonetheless, the guidelines advise that patients with inappropriately retained PD stents should undergo endoscopic removal.9-11
The data on the incidence and risk factors associated with retained PD stents remain sparse. A recent study by Martin et al.12 reported a spontaneous clearance rate of 52% and noted that the lack of an institutional protocol and the absence of meticulous tracking were barriers to adherence. In contrast, prior studies reported substantially higher spontaneous passage rates, ranging from 82.4% to 98%.13-16
The conflicting data and lack of clarity regarding the selection of an appropriate stent and the risk of prolonged stent retention may influence endoscopists’ decision-making and potentially lead to the underutilization of PD stents.12,17,18 Furthermore, the absence of a clearly defined protocol for PD stent management makes interpretation of the available literature challenging. Therefore, we performed this study to assess the frequency and risk factors associated with inappropriate retention of PD stents at 2 weeks after ERCP at a high-volume center with a clearly defined, PD stent-specific protocol for follow-up and management.
METHODS
Study design
In this retrospective, single-center study conducted at a tertiary care referral center, patients who underwent ERCP with placement of a stent into the PD were identified from a prospectively maintained and actively managed internal database. All ERCP procedures were performed between 2019 and 2024 by one of four advanced endoscopists with extensive therapeutic experience. Trainee endoscopists were not involved in the procedures. Pre-procedure administration of lactated Ringer’s solution and peri-procedural administration of indomethacin were performed unless specifically contraindicated.10,11
Database and standardized institutional protocol for stents
At our institution, all endoscopically placed biliary and pancreatic stents are systematically tracked using a prospective, secure internal database. This database facilitates longitudinal follow-up by ensuring that appropriate imaging studies and subsequent endoscopic procedures are ordered in a timely manner. Patients are expected to undergo follow-up imaging within 14 days after ERCP with placement of a PD stent. This imaging is often an abdominal radiograph but may also include computed tomography of the abdomen when clinically indicated.
If imaging is not completed within this timeframe, the clinical team initiates patient outreach to provide reminders and assist with scheduling within an additional 2 weeks. For patients who have relocated, including those who have moved out of state, efforts are made to coordinate care with their new primary care providers to ensure continuity and completion of the recommended imaging. Patients identified as having retained stents on follow-up imaging are promptly scheduled for repeat ERCP or esophagogastroduodenoscopy (EGD). A subset of patients may elect to delay endoscopic removal and instead undergo repeat imaging, particularly when radiologic evaluation is already planned for another clinical indication.
The database is actively managed by a dedicated physician assistant, with operational support from the scheduling team and oversight by the interventional endoscopy service. This structured approach is designed to ensure appropriate follow-up after ERCP and to minimize the risk of stent-related complications through proactive surveillance and coordinated care.
Inclusion and exclusion criteria
All patients who underwent ERCP with placement of plastic stents (Boston Scientific) into the PD were included in the initial assessment. We excluded patients with transgastric stents placed via endoscopic ultrasound-guided pancreatogastrostomy, lumen-apposing metal stents, pancreatic stents placed in the bile duct, as well as patients with altered foregut anatomy, including those with stents placed for PD leak after a Whipple procedure. ERCP procedures performed using a disposable duodenoscope were excluded to minimize confounding variables.
Definitions
The indication for ERCP was categorized as biliary, pancreatic, or dual based on the site of pathology and whether cannulation of the PD was deliberate (pancreatic indication) or inadvertent (biliary indication). In patients with multiple indications for ERCP, the primary indication was determined according to the predominant clinical presentation and primary billing diagnosis code associated with the procedure. Patients with concomitant biliary and pancreatic pathologies requiring intervention were defined as having a dual indication.
The intent of stent placement was categorized as prophylactic or therapeutic based on the indication for ERCP (vide supra), procedural documentation, and the type of stent selected. PD stents placed with the intent to prevent post-ERCP pancreatitis were defined as prophylactic stents. PD stents placed for the treatment of pancreatic disorders were defined as therapeutic stents.
Spontaneous passage was defined as the absence of a PD stent on imaging performed within 2 weeks. Stent retention was defined as the presence of a stent within the PD on imaging performed within 2 weeks. Inappropriate retention was defined as the presence of a non-flanged PD stent within the PD on imaging performed more than 2 weeks after ERCP.
The diameter of a PD stent is expressed in French (Fr) units and its length in centimeters. Because non-passage rates for 3 Fr and 5 Fr stents have been reported to be similar,19 these sizes were combined for binary analysis. Stents with a diameter >5 Fr were considered large-caliber stents, whereas those ≤5 Fr were considered narrow-caliber stents. Stents ≤3 cm in length were categorized as short stents, and stents >3 cm were categorized as long stents.
Outcomes
The primary analysis compared patients with retained PD stents and those with spontaneous migration. A secondary analysis evaluated outcomes according to the intent of PD stent placement—prophylactic versus therapeutic. The primary endpoint was the incidence of inappropriate retention of PD stents at 2 weeks after ERCP. Secondary endpoints included risk factors for retention, the incidence of post-ERCP pancreatitis, and the need for endoscopic removal of PD stents.
Statistical analysis
Univariate analyses were conducted using the Wilcoxon rank-sum test for continuous variables and Pearson’s chi-square test or Fisher’s exact test for categorical variables. Multivariable logistic regression with stepwise selection (both forward and backward) was used to identify significant predictors, guided by predefined criteria, including the Akaike information criterion (AIC) and p-value thresholds. Starting from the full model, predictors were iteratively added or removed to minimize the AIC until no further improvement was achieved. Descriptive statistics are reported as mean±standard deviation for continuous variables and count (percentage) for categorical variables. A two-sided p<0.05 was considered statistically significant.
Ethical statement
The study was approved by the Institutional Review Board of Providence Sacred Heart Medical Center (IRB No. 2024000209). Informed consent was waived for this retrospective study.
RESULTS
Of the 501 patients who met the selection criteria, 19 duplicate records were excluded, and the remaining 482 patients (mean age, 57.09±18.49 years; 55% female; Table 1) were included in the analysis (Fig. 1). These included 178 procedures performed in hospitalized patients.
Table 1.
Patient- and procedure-specific characteristics for pancreatic duct stenting
| Characteristic | Overall (n=482) | Retained stent (n=158) | Stent migrated (n=267) | p-value |
|---|---|---|---|---|
| Demographics | ||||
| Age (yr) | 57.09±18.49 | 57.2±18.06 | 57.51±18.11 | 0.57 |
| Female | 264 (54.8) | 79 (50.0) | 154 (57.7) | 0.13 |
| White | 400 (83.0) | 138 (87.3) | 221 (82.8) | 0.27 |
| BMI (kg/m2) | 28.71±7.3 | 27.4±6.7 | 29.28±7.5 | 0.02 |
| Inpatient | 178 (36.9) | 53 (33.5) | 104 (39.0) | 0.29 |
| Indication for ERCP | ||||
| Biliary | 289 (60.0) | 51 (32.3) | 190 (71.2) | <0.01 |
| Choledocholithiasis | 148 (30.7) | 17 (10.8) | 97 (36.3) | |
| Bile leak | 33 (6.8) | 9 (5.7) | 24 (9.0) | |
| Biliary stricture | 22 (4.6) | 6 (3.8) | 11 (4.1) | |
| Elevated liver enzymes | 25 (5.2) | 3 (1.9) | 18 (6.7) | |
| Malignant obstruction | 42 (8.7) | 12 (7.6) | 27 (10.1) | |
| Miscellaneous | 19 (3.9) | 4 (2.5) | 13 (4.9) | |
| Pancreatic | 181 (37.6) | 98 (62.0) | 74 (27.7) | |
| Chronic pancreatitis | 97 (20.1) | 58 (36.7) | 34 (12.7) | |
| Pancreatic duct leak | 38 (7.9) | 15 (9.5) | 22 (8.2) | |
| Pancreas divisum | 13 (2.7) | 8 (5.1) | 5 (1.9) | |
| Pancreatic duct stricture | 18 (3.7) | 11 (7.0) | 7 (2.6) | |
| Miscellaneous | 15 (3.1) | 6 (3.8) | 6 (2.2) | |
| Both biliary and pancreatic | 12 (2.5) | 9 (5.7) | 3 (1.1) | |
| Indication for pancreatic duct stent | <0.01 | |||
| Pancreatitis prophylaxis | 313 (64.9) | 63 (39.9) | 209 (78.3) | |
| Chronic pancreatitis | 106 (22.0) | 64 (40.5) | 36 (13.5) | |
| Pancreatic duct leak | 24 (5.0) | 13 (8.2) | 10 (3.7) | |
| Pancreatic duct stricture | 18 (3.7) | 11 (7.0) | 7 (2.6) | |
| Pancreas divisum | 12 (2.5) | 7 (4.4) | 5 (1.9) |
Values are presented as mean±standard deviation or number (%).
BMI, body mass index; ERCP, endoscopic retrograde cholangiopancreatography.
Fig. 1.
Flowchart of patient selection and outcomes. ERCP, endoscopic retrograde cholangiopancreatography; EGD, esophagogastroduodenoscopy.
Indication for ERCP
The most common indication for ERCP was choledocholithiasis (n=148, 30.7%), followed by chronic pancreatitis with intraductal stones or strictures (n=97, 20.1%). A dual biliary and pancreatic indication for ERCP was identified in 12 patients with chronic pancreatitis and pancreatolithiasis, along with concomitant extrinsic compression of the common bile duct leading to obstructive jaundice (Table 1).
Notably, the indication for ERCP was distinct from the indication for PD stent placement, as PD stents were placed during ERCP procedures performed for biliary indications when the PD was inadvertently cannulated.
Indication for PD stent
The most common indication for PD stent placement was the prevention of post-ERCP pancreatitis (n=313, 73.6%). This group included patients with a variety of pathologies, ranging from biliary diseases with inadvertent cannulation of the PD to patients with pancreatic pathology completing endoscopic therapy and initiating a stent-free trial, in whom a PD stent without an internal flange was placed to minimize the risk of post-ERCP pancreatitis. The 5 Fr and 4 Fr stents with a length of 3 cm were the most commonly used sizes (Table 2). Pancreatic stents without internal flanges were used more frequently for biliary indications than for pancreatic indications (74.4% vs. 31.5%, p<0.001).
Table 2.
Stent characteristics and outcomes after index ERCP
| Overall (n=482) | Patients with long-term follow-up (n=425) | |||
|---|---|---|---|---|
| Retained stent (n=158) | Stent migrated (n=267) | p-value | ||
| Characteristics of pancreatic duct stent | ||||
| Diameter (Fr) | <0.01 | |||
| 4 | 191 | 19 | 144 | |
| 5 | 211 | 84 | 104 | |
| 7 | 73 | 51 | 16 | |
| 10 | 2 | 1 | 1 | |
| 7+5 | 5 | 3 | 2 | |
| Length (cm) | <0.01 | |||
| 2 | 2 | 1 | ||
| 3 | 289 | 61 | 214 | |
| 4 | 16 | 5 | 8 | |
| 5 | 62 | 43 | 19 | |
| 6 | 4 | 3 | ||
| 7 | 45 | 34 | 10 | |
| 8 | 10 | 4 | 6 | |
| 9 | 20 | 11 | 6 | |
| Fluoroscopy (sec) | 64 (33–152) | 98 (39–185) | 50 (26–128) | <0.01 |
| Internal flange present | 208 | 129 | 65 | <0.01 |
| Outcomes | ||||
| Post-ERCP pancreatitis | 20 | 8 | 11 | 0.59 |
| Bleeding | 3 | 2 | 1 | 0.6 |
Values are presented as number or median (interquartile range).
ERCP, endoscopic retrograde cholangiopancreatography.
Follow-up
Of the 482 patients who underwent PD stent placement, follow-up to assess stent passage at 2 weeks was not feasible in 57 (11.8%) patients (Fig. 1). This group included patients who died before undergoing follow-up imaging, transitioned to comfort care, underwent a Whipple procedure, declined further imaging when contacted by the endoscopy team, or did not undergo imaging despite having active orders and documented contact by the clinical team.
Patients without follow-up were more likely to have a biliary indication for ERCP (n=48; 84% vs. n=241; 57%; p<0.01) and less likely to have stents with an internal flange (n=13; 23% vs. n=185; 46%; p<0.01). No significant differences were observed between groups in demographics, other stent characteristics, or fluoroscopy time.
Procedure outcomes were analyzed in the remaining 425 patients, including 158 (37.2%) with retained stents and 267 (62.8%) with spontaneous passage (Fig. 1, Tables 1, 2).
Variable selection for the logistic regression model was performed using stepwise selection based on the AIC. The full model had an AIC of 386.76, and the lowest observed AIC was 371.12. The final model included the presence of an internal flange, stent length >3 cm, diameter ≥5 Fr, pancreatic indication for ERCP, inpatient status, and sex, representing the most parsimonious set of predictors with optimal model fit. Alternative methods, including clinically guided selection and Lasso regression, yielded higher AIC values and deviance and did not improve model performance.
Retained stents
Among patients with retained PD stents on follow-up imaging (n=158), the most common indications for ERCP and for stent placement were pancreatic pathologies, including chronic pancreatitis, PD strictures, and leaks (Table 1).
Multivariable analysis demonstrated a statistically significant increase in the odds of stent retention associated with PD stents with an internal flange (odds ratio [OR], 6.39; 95% confidence interval [CI], 3.29–12.67; p<0.001; Table 3), pancreatic indication for ERCP (OR, 2.23; 95% CI, 1.22–4.07; p=0.008), and stent length >3 cm (OR, 2.5; 95% CI, 1.23–5.07; p=0.01).
Table 3.
Multivariable analysis of risk factors for retention of pancreatic duct stents at 2 weeks after index ERCP
| Parameter | OR (95% CI) | p-value |
|---|---|---|
| Stent characteristics | ||
| Presence of an internal flange | 6.39 (3.29–12.67) | <0.001 |
| Length >3 cm | 2.5 (1.23–5.07) | 0.01 |
| Diameter ≥5 Fr | 1.36 (0.62–3.01) | 0.44 |
| Patient characteristics | ||
| Pancreatic indication for ERCP | 2.23 (1.22–4.07) | 0.008 |
| Inpatient status | 0.59 (0.33–1.05) | 0.07 |
| Male sex | 0.63 (0.35–1.08) | 0.1 |
ERCP, endoscopic retrograde cholangiopancreatography; OR, odds ratio; CI, confidence interval.
Notably, in this analysis, no significant difference was observed between stent diameters >5 Fr and ≤5 Fr. This finding was attributed to confounding, as a stent length of 3 cm strongly correlated with stent diameters of 4 Fr and 5 Fr.
Therapeutic versus prophylactic stents
ERCP procedures were classified according to the intent of PD stent placement as therapeutic or prophylactic. Multivariable analysis revealed that prophylactic stents were significantly less likely to have a diameter >5 Fr (OR, 0.28; 95% CI, 0.12–0.62; p=0.002; Table 4) or a length >3 cm (OR, 0.28; 95% CI, 0.15–0.52; p=0.0001) compared with therapeutic stents. The incidence of post-ERCP pancreatitis did not differ significantly between the two groups. However, stent retention was less likely with prophylactic stents (OR, 0.46; 95% CI, 0.26–0.83; p=0.009).
Table 4.
Pancreatic duct stent characteristics and outcomes for prophylactic versus therapeutic indications
| Parameter | OR (95% CI) | p-value |
|---|---|---|
| Stent features | ||
| Diameter >5 Fr | 0.28 (0.12–0.62) | 0.002 |
| Length >3 cm | 0.28 (0.15–0.52) | 0.0001 |
| Internal flange present | 0.86 (0.39–1.92) | NS |
| Outcomes | ||
| Post-ERCP pancreatitis | 0.63 (0.17–2.33) | NS |
| Stent retention | 0.46 (0.26–0.83) | 0.009 |
OR, odds ratio; CI, confidence interval; NS, not significant; ERCP, endoscopic retrograde cholangiopancreatography.
Inappropriate stent retention
PD stents without an internal flange were placed for prophylaxis against post-ERCP pancreatitis in 231 patients, among whom inappropriate stent retention occurred in 29 (12.5%), with only 5 (2.2%) requiring endoscopic removal (Fig. 1). These five patients had a concomitant covered metal stent in the bile duct (n=3), an anomalous pancreaticobiliary junction (n=1), or a transgastric jejunal feeding tube traversing the ampulla (n=1). Removal of retained PD stents did not result in pancreatitis in any of these patients.
The remaining 24 patients included 5 who underwent repeat ERCP or EGD with the intent of stent removal; however, the PD stent was not visualized, thereby confirming spontaneous, albeit delayed, stent passage. An additional 19 patients underwent repeat imaging that demonstrated passage of the PD stent, thus obviating the need for repeat endoscopy for stent retrieval.
Secondary outcomes
Among the 482 patients who underwent ERCP, indomethacin could not be administered in 33 (6.8%). The most common reasons were a low glomerular filtration rate (n=21), a documented allergy to nonsteroidal anti-inflammatory drugs (n=4), and concern for bleeding after sphincterotomy that ultimately resolved spontaneously without intervention (n=3). The specific reason was not documented in the remaining five patients.
Overall, the incidence of post-ERCP pancreatitis during the index ERCP was 4%, with no statistically significant difference between patients with retained and spontaneously migrated stents (5.1% vs. 4%, respectively; p=0.6; Table 2). Internal migration of PD stents was not observed in this cohort.
DISCUSSION
Placement of PD stents is commonly performed to prevent post-ERCP pancreatitis or for the treatment of pancreatic disorders. Stents used for prophylaxis against post-ERCP pancreatitis are typically narrow-caliber, short, and devoid of internal flanges to facilitate spontaneous migration.8-10 However, spontaneous passage of PD stents may fail in a subset of patients. This study reports results from a tertiary care center with a standardized institutional protocol for follow-up of PD stents and demonstrates that the risk of inappropriate stent retention is very low and typically occurs due to the presence of mechanical obstruction. It also provides insights into stent characteristics and patient-related factors that influence the likelihood of stent retention and migration.
The dimensions of the ideal PD stent remain under investigation and should be individualized according to the indication for ERCP, the desired duration of stent retention, and ductal anatomy. Data suggest that a 5 Fr stent without an internal flange has a higher incidence of spontaneous migration, a lower risk of pancreatitis, and easier placement into the PD compared with a longer 3 Fr stent.9,16,19 While some studies have reported a higher risk of pancreatitis and stent retention with a 3 Fr stent,16 others have reported similar rates.19 Similarly, 5 Fr PD stents have been reported to migrate spontaneously earlier and more frequently than 7 Fr stents.20 Cannulation and stent placement success rates for 5 Fr and 3 Fr PD stents have been reported to be similar.21 Spontaneous migration has been reported to be more frequent with the 4 Fr stent than with the 5 Fr stent, although both sizes have a similar reduction in the risk of post-ERCP pancreatitis.22 PD stents longer than 8 cm are at risk of occlusion.8,23 We did not use straight-shaped stents in this study due to the risk of proximal migration.24
While short stents were the most frequently used overall, our analysis showed that longer stents were associated with higher odds of retention. Our findings further suggest that, beyond stent characteristics, therapeutic indications—such as chronic pancreatitis, ductal strictures, leaks, and stones—may contribute to increased stent retention, likely reflecting underlying pathological changes in PD morphology.1 Importantly, migration or retention of the PD stent did not appear to affect the incidence of post-ERCP pancreatitis.
In our study, inappropriately retained prophylactic PD stents that failed to migrate spontaneously were often associated with mechanical obstruction, such as the presence of a biliary stent or a feeding tube, which impeded stent passage. The presence of a biliary stent as a predictor of PD stent retention has been reported previously,12 and patients undergoing ERCP with biliary stenting should be counseled regarding the risk of PD stent retention during the informed consent process.25 Conversely, spontaneous stent migration may occur despite the use of a flanged PD stent in patients with pancreatic pathologies. Endoscopists should remain aware of the risk of spontaneous stent migration in patients with PD leaks or chronic pancreatitis with strictures, which may result in treatment failure and necessitate early diagnosis and timely reintervention.
The ideal timing for stent removal is not well defined and varies across practices.11 The timeline for stent removal is informed by data suggesting morphological changes in the PD with a retained stent,6,7 although the risk of ductal changes depends on stent size,14 the presence of an internal flange, and the material used for its construction.8,26 Furthermore, it is unclear whether these morphological changes in the PD translate into long-term clinical adverse events.27 Indeed, some endoscopists have suggested that extraction of retained stents may be performed during a planned follow-up endoscopy in selected patients.27 Notably, a subset of patients in our study with retained stents on follow-up radiography demonstrated passage of the stent on subsequent imaging or during follow-up endoscopic evaluation. This finding suggests that spontaneous passage may be delayed for a short duration but can still occur before the patient presents for endoscopic removal. Another possible explanation is misinterpretation of the imaging study by the radiologist.15 Chahal and colleagues have also reported an incremental increase in PD stent dislodgement at 2 weeks and recommended abdominal imaging at 2 weeks rather than earlier.16
This study differs from a recently published study that reported a spontaneous clearance rate of 52% at 4 weeks.12 The study by Martin et al.12 excluded patients with pancreatic pathology, whereas we included these patients as an external comparison group. Furthermore, that study was limited by wide variability in ERCP procedure volume among the 13 endoscopists and included trainee physicians. Most importantly, the study reported data from a system that lacked a standardized institutional protocol for managing PD stents and monitoring compliance and follow-up. In addition, 72 (27.5%) of the 262 analyzed patients had long stents measuring >3 cm, which is associated with an increased risk of stent retention. Finally, prophylactic indomethacin was not administered in 24% of patients in that study. In contrast, our study reports data from four highly experienced endoscopists at a high-volume referral center with a dedicated protocol for tracking patients and ensuring follow-up. Our institutional protocol requires universal administration of indomethacin and the use of narrow-caliber, short-length stents for prophylactic indications. These differences in protocol and institutional practices may explain the substantially higher incidence of retained PD stents, lower compliance with planned imaging studies, and the significantly higher incidence of post-ERCP pancreatitis reported by Martin et al.12 Other studies have also reported substantially higher rates of spontaneous PD stent migration.28
Although ERCP is a widely performed procedure for pancreaticobiliary disorders, there is limited literature specifically addressing follow-up adherence after PD stent placement. One study evaluating post-ERCP imaging for prophylactic PD stents reported a follow-up compliance rate of only 66%,15 highlighting a significant gap in surveillance. This study demonstrates improved compliance with timely follow-up imaging under a standardized protocol.
Our study is limited by its retrospective, single-center design, loss to follow-up in a subset of patients, and the associated risk of inherent biases. The generalizability of these findings may be limited in centers with lower procedural volumes or without dedicated personnel implementing a standardized protocol for follow-up and management of PD stents. Nonetheless, this study addresses a clinically relevant and underexplored question regarding the necessity of follow-up in patients with PD stents. It provides real-world data from a high-volume center with a standardized protocol.
In conclusion, this study reports that short, narrow-caliber PD stents without an internal flange placed for prophylaxis typically migrate spontaneously, and the need for endoscopic removal is uncommon. We advocate for prospective randomized controlled trials to further investigate these findings and for institutional protocols with dedicated personnel responsible for ensuring follow-up and adherence.
Footnotes
Conflicts of Interest
Divyanshoo R. Kohli received consultancy, research support, and equipment loans from Fujifilm (current) and Olympus (completed) for unrelated projects; Nishant Puri served as a consultant for Boston Scientific (completed); all other authors have no potential conflicts of interest.
Funding
None.
Acknowledgments
The authors gratefully acknowledge the assistance of Ms. Julie Campbell in conducting this study.
Author Contributions
Conceptualization: DK; Investigation: DK,NP, JB,WS; Data curation: XG; Methodology: SC; Writing–original daft: DK; Writing–review & editing: XG, JB, NP, WS, SC.
REFERENCES
- 1.Sheth SG, Machicado JD, Chalhoub JM, et al. American Society for Gastrointestinal Endoscopy guideline on the role of endoscopy in the management of chronic pancreatitis: summary and recommendations. Gastrointest Endosc. 2024;100:584–594. doi: 10.1016/j.gie.2024.05.016. [DOI] [PubMed] [Google Scholar]
- 2.Kohli DR, Shah TU, BouHaidar DS, et al. Significant infections in liver transplant recipients undergoing endoscopic retrograde cholangiography are few and unaffected by prophylactic antibiotics. Dig Liver Dis. 2018;50:1220–1224. doi: 10.1016/j.dld.2018.05.014. [DOI] [PubMed] [Google Scholar]
- 3.Kohli DR, Harrison ME, Mujahed T, et al. Outcomes of endoscopic therapy in donation after cardiac death liver transplant biliary strictures. HPB (Oxford) 2020;22:979–986. doi: 10.1016/j.hpb.2019.10.018. [DOI] [PubMed] [Google Scholar]
- 4.Sachdev A, Kashyap JR, D'Cruz S, et al. Safety and efficacy of therapeutic endoscopic interventions in the management of biliary leak. Indian J Gastroenterol. 2012;31:253–257. doi: 10.1007/s12664-012-0209-x. [DOI] [PubMed] [Google Scholar]
- 5.Kohli DR, Amateau SK, Desai M, et al. American Society for Gastrointestinal Endoscopy guideline on management of post-liver transplant biliary strictures: summary and recommendations. Gastrointest Endosc. 2023;97:607–614. doi: 10.1016/j.gie.2022.10.007. [DOI] [PubMed] [Google Scholar]
- 6.Sherman S, Hawes RH, Savides TJ, et al. Stent-induced pancreatic ductal and parenchymal changes: correlation of endoscopic ultrasound with ERCP. Gastrointest Endosc. 1996;44:276–282. doi: 10.1016/s0016-5107(96)70164-5. [DOI] [PubMed] [Google Scholar]
- 7.Kozarek RA. Pancreatic stents can induce ductal changes consistent with chronic pancreatitis. Gastrointest Endosc. 1990;36:93–95. doi: 10.1016/s0016-5107(90)70958-3. [DOI] [PubMed] [Google Scholar]
- 8.Han S, Obando JV, Bhatt A, et al. Biliary and pancreatic stents. Gastrointest Endosc. 2023;97:1003–1004. doi: 10.1016/j.gie.2023.03.016. [DOI] [PubMed] [Google Scholar]
- 9.Dumonceau JM, Kapral C, Aabakken L, et al. ERCP-related adverse events: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy. 2020;52:127–149. doi: 10.1055/a-1075-4080. [DOI] [PubMed] [Google Scholar]
- 10.Buxbaum JL, Freeman M, Amateau SK, et al. American Society for Gastrointestinal Endoscopy guideline on post-ERCP pancreatitis prevention strategies: summary and recommendations. Gastrointest Endosc. 2023;97:153–162. doi: 10.1016/j.gie.2022.10.005. [DOI] [PubMed] [Google Scholar]
- 11.Buxbaum JL, Freeman M, Amateau SK, et al. American Society for Gastrointestinal Endoscopy guideline on post-ERCP pancreatitis prevention strategies: methodology and review of evidence. Gastrointest Endosc. 2023;97:163–183. doi: 10.1016/j.gie.2022.09.011. [DOI] [PubMed] [Google Scholar]
- 12.Martin J, Alikhanova A, Srikureja W, et al. Frequency and predictors of delayed clearance of prophylactic pancreatic stents after ERCP. Gastrointest Endosc. 2025;102:845–851. doi: 10.1016/j.gie.2025.03.638. [DOI] [PubMed] [Google Scholar]
- 13.Moffatt DC, Coté GA, Fogel EL, et al. Acute pancreatitis after removal of retained prophylactic pancreatic stents. Gastrointest Endosc. 2011;73:980–986. doi: 10.1016/j.gie.2011.01.012. [DOI] [PubMed] [Google Scholar]
- 14.Rashdan A, Fogel EL, McHenry L, et al. Improved stent characteristics for prophylaxis of post-ERCP pancreatitis. Clin Gastroenterol Hepatol. 2004;2:322–329. doi: 10.1016/s1542-3565(04)00062-x. [DOI] [PubMed] [Google Scholar]
- 15.Loloi J, Lipkin JS, Gagliardi EM, et al. Assessing spontaneous passage of prophylactic pancreatic duct stents by X-ray: is a radiology report adequate? Ther Adv Gastrointest Endosc. 2019;12:2631774519862895. doi: 10.1177/2631774519862895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chahal P, Tarnasky PR, Petersen BT, et al. Short 5Fr vs long 3Fr pancreatic stents in patients at risk for post-endoscopic retrograde cholangiopancreatography pancreatitis. Clin Gastroenterol Hepatol. 2009;7:834–839. doi: 10.1016/j.cgh.2009.05.002. [DOI] [PubMed] [Google Scholar]
- 17.Freeman ML, Overby C, Qi D. Pancreatic stent insertion: consequences of failure and results of a modified technique to maximize success. Gastrointest Endosc. 2004;59:8–14. doi: 10.1016/s0016-5107(03)02530-6. [DOI] [PubMed] [Google Scholar]
- 18.Avila P, Holmes I, Kouanda A, et al. Practice patterns of post-ERCP pancreatitis prophylaxis techniques in the United States: a survey of advanced endoscopists. Gastrointest Endosc. 2020;91:568–573. doi: 10.1016/j.gie.2019.11.013. [DOI] [PubMed] [Google Scholar]
- 19.Zolotarevsky E, Fehmi SM, Anderson MA, et al. Prophylactic 5-Fr pancreatic duct stents are superior to 3-Fr stents: a randomized controlled trial. Endoscopy. 2011;43:325–330. doi: 10.1055/s-0030-1256305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chahal P, Baron TH, Petersen BT, et al. Pancreatic stent prophylaxis of post endoscopic retrograde cholangiopancreatography pancreatitis: spontaneous migration rates and clinical outcomes. Minerva Gastroenterol Dietol. 2007;53:225–230. [PubMed] [Google Scholar]
- 21.Ramesh J, Kim H, Reddy K, et al. Impact of pancreatic stent caliber on post-endoscopic retrograde cholangiopancreatogram pancreatitis rates in patients with confirmed sphincter of Oddi dysfunction. J Gastroenterol Hepatol. 2014;29:1563–1567. doi: 10.1111/jgh.12585. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pahk A, Rigaux J, Poreddy V, et al. Prophylactic pancreatic stents: does size matter?: a comparison of 4-Fr and 5-Fr stents in reference to post-ERCP pancreatitis and migration rate. Dig Dis Sci. 2011;56:3058–3064. doi: 10.1007/s10620-011-1695-x. [DOI] [PubMed] [Google Scholar]
- 23.Farnbacher MJ, Radespiel-Tröger M, König MD, et al. Pancreatic endoprostheses in chronic pancreatitis: criteria to predict stent occlusion. Gastrointest Endosc. 2006;63:60–66. doi: 10.1016/j.gie.2005.08.046. [DOI] [PubMed] [Google Scholar]
- 24.Cai YL, Wang F, Zhang J, et al. Incidence, risk factors and endoscopic retrieval of proximally migrated pancreatic stents: 25 years of experience in a high-volume center - a case controlled study. Int J Surg. 2025;111:3115–3119. doi: 10.1097/js9.0000000000002273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Storm AC, Fishman DS, Buxbaum JL, et al. American Society for Gastrointestinal Endoscopy guideline on informed consent for GI endoscopic procedures. Gastrointest Endosc. 2022;95:207–215. doi: 10.1016/j.gie.2021.10.022. [DOI] [PubMed] [Google Scholar]
- 26.Samuelson A, Zeligman B, Russ P, et al. pancreatic duct changes in patients with chronic pancreatitis treated with polyethylene and Sof-Flex material stents: a blinded comparison. Pancreas. 2016;45:281–285. doi: 10.1097/mpa.0000000000000471. [DOI] [PubMed] [Google Scholar]
- 27.Dultz G, Gerber L, Zeuzem S, et al. Prolonged retention of prophylactic pancreatic stents is not associated with increased complications. Pancreatology. 2019;19:39–43. doi: 10.1016/j.pan.2018.11.011. [DOI] [PubMed] [Google Scholar]
- 28.Lawrence C, Cotton PB, Romagnuolo J, et al. Small prophylactic pancreatic duct stents: an assessment of spontaneous passage and stent-induced ductal abnormalities. Endoscopy. 2007;39:1082–1085. doi: 10.1055/s-2007-966815. [DOI] [PubMed] [Google Scholar]