A stepwise cannulation strategy for conservative endoscopists: the clinical impact of transpancreatic precut after pancreatic stenting in a retrospective study from Taiwan
Article information
Abstract
Background/Aims
Pancreatic stenting reduces post-endoscopic retrograde cholangiopancreatography pancreatitis (PEP) and aids in cannulation in difficult cases. However, conservative endoscopists may stop at this step, resulting in suboptimal outcomes. This study assessed the efficacy of transpancreatic precut sphincterotomy (TPS) as a rescue procedure following pancreatic stenting.
Methods
Between March 2013 and November 2018, 82 patients underwent pancreatic stenting at our institution prior to successful biliary cannulation. TPS was introduced in April 2016, and patients were divided into Before TPS and After TPS groups. The outcomes included cannulation success, PEP incidence, and predictors of TPS conversion.
Results
There were 43 and 39 patients in the Before TPS and After TPS groups, respectively. Twenty-two patients (56.4%) underwent conversion to TPS in the After TPS group. The After TPS group had a higher bile duct cannulation rate (89.7% vs. 72.1%) than the Before TPS group, but this difference was not statistically significant (p=0.054). Multivariate analysis showed that age >50 years (odds ratio [OR], 0.181; p=0.021) and being in the After TPS group (OR, 0.712; p=0.039) were independently associated with reduced PEP risk. Haraldsson Type 2 and Type 4 papillae carried a relatively high TPS conversion rate.
Conclusions
A stepwise cannulation strategy that incorporates TPS after pancreatic stenting minimizes the need for advanced techniques and improves PEP outcomes.
INTRODUCTION
Endoscopic retrograde cholangiopancreatography (ERCP) is a widely used procedure for diagnosing and treating biliary tract disorders, including stone removal, stenting, and tissue sampling. However, ERCP is associated with a high risk of complications, particularly post-ERCP pancreatitis (PEP), which is strongly associated with prolonged cannulation attempts1 and unintended pancreatic duct cannulation.2 Therefore, the development of effective cannulation strategies is essential for optimizing procedural success while minimizing complications.
Unintended pancreatic duct cannulation is common during ERCP, particularly in cases of difficult biliary cannulation.2 To reduce the risk of PEP in such settings, the European Society of Gastrointestinal Endoscopy (ESGE) recommends the placement of a prophylactic pancreatic stent.3 The ESGE also endorses the use of advanced cannulation techniques, such as the double-guidewire technique (DGT), needle-knife precut (NKP), and transpancreatic precut sphincterotomy (TPS), when standard methods fail. Although these techniques have demonstrated efficacy in improving biliary cannulation rates, their technical complexity and increased complication risk4,5 often limit their adoption, especially among less experienced6,7 or more conservative endoscopists. In real-world practice, some procedures may be prematurely terminated after pancreatic stent placement if cannulation remains unsuccessful, potentially missing an opportunity for definitive therapy.
Among the available rescue options, the pancreatic stent-assisted technique, which is performed via a standard cannulation maneuver alongside a pancreatic stent, is often favored because of its relative simplicity. First described in 1996,8 this technique offers several advantages. The pancreatic stent can straighten the pancreatic duct and common channel, stabilize any papillae, and widen the papillary orifice, thereby facilitating biliary access and potentially reducing the need for precut techniques.9 However, previous studies have reported suboptimal outcomes in terms of success rates and complication profiles.10 When the pancreatic stent-assisted technique fails, TPS is the next logical step. TPS mimics standard wire-guided sphincterotomy but is performed over a pancreatic duct guidewire and has been shown to be technically feasible with a favorable safety profile, even in the hands of less-experienced operators.11 Therefore, TPS may serve as an effective and accessible rescue strategy, allowing endoscopists to maintain a stepwise approach without resorting to more technically demanding or high-risk methods.
At our institution, minimizing ERCP-related complications has been a top priority. In cases of difficult cannulation, prophylactic pancreatic stenting is commonly used, and the pancreatic stent-assisted technique is routinely employed as the first-line advanced approach. In April 2016, TPS was introduced and has since become the preferred rescue strategy when stent-assisted cannulation fails because of its technical feasibility and favorable safety profile. This study aimed to evaluate the clinical impact of incorporating TPS as a rescue technique following unsuccessful pancreatic stent-assisted biliary cannulation in patients with unintended pancreatic duct access. By comparing outcomes before and after TPS became part of our standard approach, we sought to determine whether TPS improved success rates while maintaining patient safety, particularly for endoscopists who favor stepwise and conservative strategies.
METHODS
Patient data collection
To assess the impact of TPS as a rescue strategy following the use of the pancreatic stent-assisted technique, we conducted a retrospective analysis of patients who had not previously undergone ERCP and who underwent ERCPs for biliary indications between March 2013 and November 2018 at our institution. During this period, prophylactic pancreatic stenting was the primary strategy for PEP prevention, whereas pharmacological prophylaxis was not routinely used until November 2018.
Patients were included if they had a pancreatic stent placed before any further attempt at biliary cannulation. ERCP procedures involving advanced cannulation techniques other than TPS or those in which pharmacological prophylaxis for PEP was used were excluded from the analysis. The collected clinical data included age, sex, relevant medical history, ERCP findings, and procedural outcomes. The ERCP indications were categorized into four groups: biliary stones, distal common bile duct obstruction, proximal bile duct obstruction, and other indications. The macroscopic morphology of the ampulla of Vater was classified into four types according to the criteria proposed by Haraldsson et al.12: regular papillae (type 1), small papillae (type 2), enlarged or protruding papillae (type 3), and creased or ridged papillae (type 4) (Fig. 1). The presence of periampullary diverticulum was also documented and categorized based on the classification system described by Shi et al. in 2023.13 Both papillary and diverticular classifications were determined retrospectively by reviewing the archived duodenoscopic images. To ensure classification accuracy, two experienced endoscopists (W.C.S. and T.H.H.) independently reviewed and categorized the images. When these endoscopists disagreed, a third endoscopist (J.H.C.) adjudicated the final classification through consensus.
(A) Type 1. Regular papilla, with no distinctive features. (B) Type 2. Small papilla, often flat, with a diameter not greater than 3 mm (approximately 9 Fr). (C) Type 3. Protruding or pendulous papilla. The papilla is standing out, protruding or bulging into the duodenal lumen or sometimes hanging down, pendulous, with the orifice oriented caudally. (D) Type 4. Creased or ridged papilla, where the ductal mucosa seems to extend distally out of the orifice either on a ridge or in a crease.
ERCP procedures
All ERCP procedures were initiated using standard cannulation techniques under monitored anesthesia. The procedures were performed by one of the four experienced ERCP endoscopists at our institution, each with 18, 9, 6, and 5 years of experience at the time of data collection (endoscopists A, B, C, and D, respectively).
Prophylactic pancreatic stenting was considered for PEP prevention in cases involving more than one unintended pancreatic duct cannulation. The timing of stent placement was left to the discretion of the endoscopist. Following pancreatic stent placement, attempts at biliary cannulation were continued until success was achieved or until further attempts were deemed unfeasible. Before April 2016, advanced cannulation techniques were rarely used at our center, with only 10 NKP and 3 DGT procedures performed among 359 ERCPs. In April 2016, TPS was introduced as a rescue technique, and its use was determined by the performing endoscopist. TPS was performed as follows: In patients in whom selective bile duct cannulation could not be achieved after pancreatic stent placement, a guidewire was advanced deep into the pancreatic duct. Wire-guided sphincterotomy was performed using a standard sphincterotome oriented in the 11 o’clock direction relative to the pancreaticobiliary septum. Following the incision, further attempts to perform bile duct cannulation were made on the left side of the sphincterotomy site.
Definitions and data measurement
The included ERCP procedures were categorized into two groups based on whether TPS was available as a rescue technique: Before TPS (procedures performed before April 2016) and After TPS (procedures performed from April 2016 onward) groups. Data collected included the timing of pancreatic stenting initiation, bile duct cannulation time, total ERCP duration, and post-ERCP adverse events. According to the consensus of Cotton et al.,14 PEP was defined as clinical pancreatitis with a serum lipase level at least three times the upper limit of normal more than 24 hours after the procedure that required a prolongation of planned hospital admission. Cholangitis was defined as a body temperature exceeding 38 °C attributed to a biliary cause without evidence of acute cholecystitis. Bleeding was defined as “immediate” if it was during the procedure or immediately afterward, and as “delayed” if it occurred hours to several weeks after the procedure. ERCP-related perforations are limited to sphincterotomy-related ones, which are identified by the presence of extraluminal air on radiological imaging. At our institution, endoscopic and fluoroscopic images are routinely captured during key procedural milestones, including initial papillary visualization, stent placement, successful cannulation, and procedure completion. The timing of pancreatic stent placement, biliary cannulation, and total ERCP duration was retrospectively determined based on timestamped image records, with the moment of initial papilla visualization used as the starting point.
Statistical analyses
Categorical variables are presented as absolute numbers with corresponding percentages, whereas continuous data are reported as means with standard deviations. For between-group comparisons, the chi-squared test or Fisher exact test was performed as appropriate for categorical variables, whereas the Student t-test was performed for continuous variables. Logistic regression analysis was performed to identify the factors associated with PEP. All statistical analyses were performed using the IBM SPSS ver. 27.0 for Windows (IBM Corp.). The reported p-values were two-tailed, and statistical significance was set at p<0.05.
Ethical considerations
The study was approved by the Institutional Review Board of Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation (14-IRB049).
RESULTS
Between March 2013 and November 2018, 920 ERCP procedures for biliary diseases were performed at our center, of which 668 were performed in patients with naive papillae. Among these, 108 underwent prophylactic pancreatic stenting to prevent PEP. Eighteen patients were excluded because advanced cannulation techniques were used before pancreatic stenting (TPS, 10; NKP, 5; and DGT, 3). Additionally, five patients underwent NKPs after pancreatic stenting, two underwent prophylactic pancreatic stenting after successful bile duct cannulation, and two received pharmacological prophylaxis for PEP. Finally, 82 ERCP procedures met the inclusion criteria and were included in the final analysis. The data collection process is illustrated in Fig. 2. Endoscopist A performed 58 procedures; the remaining 24 procedures were almost evenly distributed among endoscopists B, C, and D. As summarized in Table 1, 43 patients (41.5%) were male, and 66 (80.5%) were over 50 years of age. The most common ERCP indication was biliary stones (87.8%), followed by distal (8.5%) and proximal (1.2%) biliary obstructions; two cases involved bile leakage. Type 1 papillae were the most common (43.9%), followed by Types 3 (22.0%), 4 (20.7%), and 2 (13.4%). Ten patients had a history of biliary pancreatitis, and two had end-stage renal disease. Difficult cannulation (cannulation time >10 minutes) occurred in 86.6% of the patients. Among the 82 procedures, 43 were performed in the After TPS group and 39 in the Before TPS group. There were no significant differences in baseline characteristics between the two groups.
Flow chart of data collection. ERCP, endoscopic retrograde cholangiopancreatography; TPS, transpancreatic precut sphincterotomy.
Characteristics of patients undergoing ERCP with continued biliary cannulation after pancreatic stent placement before and after TPS introduction
As shown in Table 2, TPS was employed as a rescue technique in 22 patients whose procedures were included in the After TPS group. There were no significant differences in bile duct cannulation times or total ERCP durations between the two groups. However, pancreatic stenting was initiated significantly earlier in the After TPS group than in the Before TPS group (9.13±5.53 vs. 15.91±11.18 minutes, p=0.001). Although the After TPS group demonstrated a higher bile duct cannulation success rate (Before TPS, 72.1%; After TPS, 89.7%; p=0.054) and a lower PEP rate (Before TPS, 20.9%; After TPS, 5.1%; p=0.051), the difference was not significant according to Fisher exact test. No statistically significant differences were observed between the two groups in terms of bleeding complications or cholangitis. No perforations were observed during any procedure.
Outcomes of patients undergoing ERCP with continued biliary cannulation after pancreatic stent placement before and after TPS introduction
Table 3 presents the results of the univariate and multivariate logistic regression analyses for the risk factors associated with PEP. According to the univariate analysis, the older (age >50 years) group had a lower risk of developing PEP (odds ratio [OR], 0.220; p=0.028). The other patient-related and procedural variables were not significantly associated with PEP. Multivariate logistic regression analysis revealed that age >50 years (OR, 0.181; 95% confidence interval [CI], 0.012–0.774; p=0.021) and inclusion in the After TPS group (OR, 0.172; 95% CI, 0.032–0.918; p=0.039) were significantly associated with a reduced incidence of PEP.
Univariate and multivariate analyses of the factors associated with post-ERCP pancreatitis
As shown in Table 4, 22 of the 39 ERCP procedures (56.4%) required conversion from the pancreatic stent-assisted technique to TPS. A significant difference in the conversion rate was observed among patients with different macroscopic papilla appearances (p=0.034). Type 4 (85.7%) and type 2 (83.3%) papillae showed the highest conversion rates, followed by type 3 (58.3%) and type 1 (28.6%) papillae. No significant differences were found in diverticulum type or ERCP indications.
Conversion rates and biliary cannulation success after TPS introduction
DISCUSSION
In this study, the addition of TPS as a rescue technique following a pancreatic stent-assisted approach significantly reduced the incidence of PEP. Moreover, macroscopic papillary morphology was significantly associated with conversion to TPS, with Types 2 and 4 papillae showing the highest conversion rates.
Among the techniques used for difficult biliary cannulation, the pancreatic stent-assisted approach is technically simple, requires no additional accessories, and can be performed using standard cannulation methods, making it particularly suitable for less-experienced endoscopists or trainees. Avoiding complex guidewire manipulation or cutting may theoretically reduce procedure-related trauma. Although the success rate of bile duct cannulation is relatively low, and the associated PEP incidence is relatively high,10 this approach offers a practical and accessible entry point for trained endoscopists. Despite the well-established influence of procedural volume and experience on ERCP outcomes, most studies evaluating advanced cannulation techniques have been conducted by expert operators, despite the well-established influence of procedural volume and experience on ERCP outcomes.7,15 For those less familiar with freehand NKP or DGT, the pancreatic stent-assisted method is a reasonable first-line strategy. Prior studies have also suggested that it may reduce the immediate need for more complex techniques.9
However, our data revealed that the pancreatic stent-assisted technique alone resulted in a cannulation success rate of only 72.1% and a PEP rate as high as 20.9%, both of which are below the ESGE-recommended quality indicators of ≥90% for cannulation success and ≤6% for PEP incidence.16 In this context, TPS serves as an effective and practical next-step rescue method. When standard wire-guided sphincterotomy is performed after guidewire insertion into the pancreatic duct, TPS is technically easier when a pancreatic stent has already been placed. In our study, the sequential use of pancreatic stenting followed by TPS improved the biliary cannulation success rate to 89.7% and reduced the incidence of PEP to 5.1%, thus meeting the guideline standards without introducing additional complexity.
Although TPS was first introduced in 1995,17 it remains relatively underutilized and is assigned a lower priority among ESGE-recommended techniques.3 In addition to being relatively understudied, TPS is a concern regarding the long-term outcomes of pancreatic sphincterotomy. However, a large-scale study involving 369 patients over an 11-year period18 and a long-term follow-up study of 134 patients over 4–10 years (median follow-up, 6 years)19 reported no late complications related to TPS. In our study, TPS was not used as a first-line method but rather as a rescue technique after pancreatic stent-assisted cannulation. This strategy not only maintained a high success rate but also limited the overall use of TPS, with a conversion rate of approximately 60% in the After TPS group. Macroscopic papillary morphology is a key predictor of conversion, with type 2 and 4 papillae showing the highest likelihood of requiring TPS. For these subtypes, early consideration of TPS, particularly after repeated unintended pancreatic cannulations, may increase procedural efficiency and safety. Nevertheless, our study did not include long-term follow-up; thus, we could not directly evaluate the long-term safety of TPS in our cohort. Although prior evidence suggests favorable long-term outcomes, future prospective studies with extended follow-up periods are needed to confirm its safety profile in diverse clinical settings.
Because radiological and surgical alternatives are available for biliary decompression, some endoscopists may adopt a more conservative strategy and refrain from pursuing technically advanced endoscopic interventions owing to the perceived procedural risks. However, premature termination after pancreatic stent placement alone may not ensure patient safety and may result in missed opportunities for definitive endoscopic treatment. A stepwise strategy that incorporates pancreatic stent-assisted cannulation, followed by TPS as a rescue method, is a practical and effective alternative. This approach increases the likelihood of successful biliary access while avoiding more complex and high-risk techniques, such as freehand NKP or DGT. This sequential approach improves the procedural success and reduces the risk of PEP by combining the protective benefits of pancreatic stenting with the technical accessibility of TPS. Importantly, the role of TPS should not be overlooked after failed pancreatic stent-assisted cannulation. Underutilization of this technique may result in avoidable therapeutic failure and compromised patient outcomes. Future prospective, randomized controlled studies directly comparing standard precut sphincterotomy and TPS are warranted to validate these findings and clarify their respective roles in the management of difficult biliary cannulation.
This study had several limitations. First, as this was a retrospective, single-center study, our analysis was inherently limited by potential selection bias. The decision to perform pancreatic duct stenting and proceed to TPS was made at the discretion of the individual endoscopist, which could have influenced the procedural outcomes. Notably, TPS was applied in less than 60% of patients in the After TPS group, indicating that the decision remained selective rather than systematic. This selective application may reflect real-world practice patterns, particularly among conservative endoscopists; however, it also underscores the need for future prospective studies with standardized decision-making criteria to reduce this bias. Second, pancreatic duct stent placement tended to be earlier in the After TPS group than in the Before TPS group, possibly reflecting a more proactive approach facilitated by the availability of TPS as a rescue technique. However, according to logistic regression analyses, this factor was not significantly associated with the PEP rate. Third, in line with the current guidelines recommending pancreatic stent placement after more than one unintended pancreatic duct cannulation, our study categorized patients based on whether they underwent more than one such cannulation without recording the exact number of pancreatic duct cannulations. In clinical practice, unsuccessful pancreatic stent-assisted cannulation is often attributed to unintended pancreatic duct cannulation during repeated attempts. This may partially explain the limited protective effect of pancreatic stenting against PEP observed in the Before TPS group because multiple attempts at pancreatic stent-assisted cannulation may have led to cumulative pancreatic duct trauma. The optimal timing for transitioning to TPS warrants further investigation in future studies.
In conclusion, the addition of TPS after pancreatic stent-assisted cannulation reduces the risk of PEP. Papillary morphology may help guide its timely application. This stepwise strategy minimizes the need for advanced techniques and is well-suited for conservative endoscopists to manage difficult biliary cannulation.
Notes
Conflicts of Interest
The authors have no potential conflicts of interest.
Funding
None.
Author Contributions
Conceptualization: WCS, CCW, THH, HDC, JHC; Data curation: WCS, THH, HDC, JHC; Formal analysis: WCS, HDC; Investigation: WCS, THH; Methodology: WCS, JHC, HDC; Resources: WCS, THH, HDC, JHC; Supervision: CCW, JHC; Validation: HDC, JHC; Visualization: WCS, CCW; Writing–original draft: WCS, THH, THK, CHC; Writing–review & editing: all authors.