Tractogastrostomy: A Case Series and Review of Novel Endoscopic Ultrasound-Guided Modality for Internalization of the External Pancreatic Fistula in the Disconnected Pancreatic Duct Syndrome

Abstract

Disconnected pancreatic duct syndrome (DPDS) often complicates acute necrotizing pancreatitis (ANP) and can result in persistent external pancreatic fistulae (EPFs), leading to significant morbidity. Traditional surgical approaches are effective, but are associated with a high risk. Endoscopic ultrasound (EUS)-guided tractogastrostomy has emerged as a minimally invasive technique to internalize EPF by creating a physiological drainage route into the stomach. This study evaluated the efficacy, safety, and technical variations of EUS-guided tractogastrostomy for DPDS-related EPF. We retrospectively analyzed 21 patients with persistent EPF secondary to DPDS who underwent EUS-guided tractogastrostomy using different techniques: Direct tract puncture, artificial fluid collection creation, wire localization, water instillation, double-scope, direct percutaneous catheter drainage (PCD) puncture, EUS-guided pancreatogastrostomy (EUS-PG), and interventional radiologist-guided rendezvous approach. The outcomes included technical success, fistula closure, complications, and follow-up data. Technical and clinical success was achieved in all patients (100%). Complete EPF closure occurred in all the patients within 3 months. Minor complications included abdominal pain (n = 4), minor bleeding (n = 1), and stent migration (n = 3, 14.28%). No major adverse events or recurrences were reported over the 6-month follow-up period. EUS-guided tractogastrostomy is a safe, effective, and minimally invasive procedure for internalizing EPF in DPDS with promising clinical outcomes.

Keywords

EUS-guided tractogastrostomy - acute necrotizing pancreatitis - disconnected pancreatic duct syndrome - pancreatitis

Introduction

Disconnected pancreatic duct syndrome (DPDS) is characterized by circumferential interruption of ductal integrity, resulting in necrosis of pancreatic ductal epithelial cells. This can lead to a partial or complete disruption of the duct. After duct disruption, the upstream viable segment secretes pancreatic juices and precipitates an external pancreatic fistula (EPF), which may last for months to years.[1] EPF poses a significant clinical challenge due to persistent external drainage, nutritional compromise, risk of infection, and impact on the patient's quality of life. The various treatment approaches for the internalization of EPF are conservative, endoscopic, and surgical. Surgical approaches, such as distal pancreatectomy, are effective but carry a high risk of morbidity and mortality.[2] Among these, endoscopic ultrasound (EUS)-guided tractogastrostomy has emerged as a novel, minimally invasive approach for internalization of EPFs. By leveraging EUS to identify the disrupted pancreatic segment or fistulous tract and establish internal drainage into the stomach, this technique offers the promise of physiological internalization, faster fistula closure, reduced dependence on external drains, and improved overall outcomes.[3] In this review, we aim to comprehensively summarize the technical advances, clinical applications, outcomes, and limitations of EUS-guided tractogastrostomy in the management of EPF in DPDS. Additionally, we explore the various techniques described in the literature, the need for standardization, and critical questions that remain unanswered in this emerging modality.

Materials and Methods

Retrospective data were collected at the Institute of Digestive and Hepatobiliary Sciences, Medanta – The Medicity, Gurugram, from January 2022 to January 2025, in patients with persistent EPF in cases of acute necrotizing pancreatitis (ANP) with DPDS. All patients were diagnosed with ANP based on the revised Atlanta Classification. A step-up approach was adopted for the management of fluid collection, which involved initial conservative treatment followed by percutaneous catheter drainage (PCD) or endoscopic drainage. The indications for collection drainage were suspected or proven infection, sepsis, persistent organ failure, and symptoms due to the compression of adjacent organs. Patients who failed to improve underwent upgrading (up to 32 Fr) or additional PCD placement. Patients who did not improve with the above treatment were considered for percutaneous endoscopic necrosectomy or surgical necrosectomy, depending on the clinical indication. A flow chart ([Fig. 1]) shows patient numbers across the acute pancreatitis (AP)–ANP–DPDS–EPF pathway. All the patients with persistent amylase-rich fluid drain output of >50 mL/day for >6 weeks were included in the study. DPDS was diagnosed based on the following criteria: (1) At least 2 cm of necrosis in the pancreas, (2) viable pancreatic tissue upstream from the site of necrosis (i.e., toward the pancreatic tail), and (3) extravasation or total cutoff of contrast material injected from the main pancreatic duct (MPD) on endoscopic retrograde cholangiopancreatography (ERCP), (4) the pancreatic duct (PD) entered the collection at an angle of approximately 90 degrees (if visible).[4] The diagnosis of DPDS was mainly made on magnetic resonance cholangiopancreatography (MRCP) imaging. ERCP was performed primarily to confirm ductal anatomy, assess MPD cutoff/extravasation, and attempt transpapillary bridging where feasible. MRCP was performed in all patients; however, due to MRCP's limitation in identifying subtle ductal cutoff or active fistula, ERCP was still required in a subset of patients. Thoroughly written informed consent was obtained from the patients before each procedure, according to the institutional protocol. All the patients received preprocedural prophylactic intravenous antibiotics. All procedures were performed by experienced endoscopists in the left-lateral position under moderate sedation with propofol. EUS was performed using a linear array echo-endoscope (GF UCT 180; Olympus Ltd., Tokyo, Japan) connected to an ultrasound unit processor (EUS ME2 Premium plus, Olympus Optical Co., Ltd., Tokyo, Japan). A 19G needle (Expect 19G, Boston Scientific, Marlborough, MA), 0.025-inch guidewire (VisiGlide Guidewire; Olympus Corp., Japan), and 6F cystotome (Cysto-Gastro-Set; ENDO-FLEX, GmbH, Voerde, Germany) were used for all the punctures.

Fig. 1 Flow chart showing the number of patients across the AP–ANP–DPDS–EPF pathway. AP, acute pancreatitis; ANP, acute necrotizing pancreatitis; DPDS, disconnected pancreatic duct syndrome; EPF, external pancreatic fistula.

Criteria Used to Determine Treatment Success and Failure

Treatment Success

Complete cessation of drainage of cutaneous pancreatic drainage and successful closure of EPF with removal of PCD within 12 weeks following EUS-guided tractogastrostomy.

Persistent Drain Output

Amylase-rich fluid output >50 mL/day persisting beyond 6 weeks despite standard step-up management.

Treatment Failure

Persistent EPF output beyond 12 weeks needs surgical intervention.

Various Procedural Techniques Used for Tractogastrostomy and Their Pros and Cons ([Table 1])

Endoscopic Ultrasound-Guided Direct Puncture Techniques

• A. Direct puncture of the fistulous tract

  This technique is used when we can localize the fistulous tract. After localizing the tip on EUS, a transgastric puncture was performed using a 19G needle. Thereafter, the guidewire was passed via the needle through the fistulous tract. Subsequently, the transmural tract was dilated with a 6 Fr cystotome, and a single 7 Fr double pigtail stent was inserted to achieve drainage of the PD ([Fig. 2]).

• B. Wire localization technique

  When the tip of the pigtail catheter is not localized on EUS, a guidewire may be passed through the PCD to appear as an echogenic landmark on the EUS. After localization, EUS-guided transgastric puncture with a 19G needle was performed, and the guidewire was advanced along the pre-established path after dilating the tract with a 6F cystotome. Later, a 5 Fr double pigtail stent was inserted, and drainage of the PD was achieved. The initial guidewire inserted via the PCD served exclusively as a visual reference to facilitate accurate alignment and guidance of the subsequent guidewire introduction after dilation of the tract ([Fig. 3]).

• C. Direct percutaneous catheter drainage puncture technique ([Supplementary Video S1])

  The PCD was localized on EUS as two echogenic lines. After localization, a transgastric puncture was performed, and the needle was placed into the PCD. The guidewire was advanced along the path after dilating the tract with a 6F cystotome. Later, a 5 Fr double pigtail stent was inserted, and drainage of the PD was achieved. The dye was instilled through the needle and localized at the PCD site, which confirms the location. Then, the 5 Fr double pigtail stent was placed transmurally for drainage of the PD ([Fig. 4]).

Fig. 2 Direct fistula tract puncture technique. Blue arrow: Guidewire through 19G needle.

Fig. 3 Air bubble localization technique. Blue arrow: Sterile water instillation through PCD. PCD, percutaneous catheter drainage.

Fig. 4 Double-scope technique. Blue arrow: Passage of the second endoscope through the external pancreatic fistula site.

Download Video

Supplementary Video S1 Direct percutaneous catheter drainage (PCD) puncture technique.

Water Instillation Techniques

• A. Air bubble localization technique

  This technique was used when the PD or PCD could not be visualized using an echoendoscope. Around 100 to 300 mL of sterile water was injected through the PCD. Air bubbles were localized on EUS, and the tract was punctured using a 19G needle. Thereafter, the guidewire was passed through the needle into the fistulous tract, the tract was dilated with a 6 Fr cystotome, and a single 7 Fr double pigtail stent was inserted to achieve internalization of the EPF ([Fig. 5]).

• B. Artificial fluid collection creation technique

  This technique was used when the PD or PCD was not visualized on the echoendoscope. This technique is based on the principle of artificial creation of pancreatic fluid collection (PFC) after clamping the PCD for 24 to 48 hours. This artificial fluid collection was localized and targeted using a linear echoendoscope, punctured using a 19G needle, and the tract dilated with a 6F cystotome. Thereafter, transgastric drainage was achieved with a 5 or 7 Fr double pigtail stent ([Fig. 6]).

Fig. 5 Wire localization technique. Blue arrow: Guidewire insertion through PCD. PCD, percutaneous catheter drainage.

Fig. 6 Direct PCD puncture. Red arrow: 19G needle. Green arrow: PCD appearing as echogenic lines. PCD, percutaneous catheter drainage.

Hybrid Techniques

• A. Double scope technique

  This procedure was performed by two experienced endoscopists. The PCD was identified on EUS as having two echogenic lines. Following localization, a transgastric puncture was performed, and the needle was positioned in proximity to the PCD. A guidewire was introduced through the needle and maneuvered to exit the PCD site. The exit of the guidewire at the fistula site was secured, and the entire tract from the gastric puncture site to the external fistula site was dilated using a 6F cystotome. In selected cases, balloon dilation (6–8 mm) was additionally performed to facilitate placement of two plastic stents. Subsequently, a 10 Fr × 5 cm double pigtail plastic stent was placed between the gastric lumen and the tract. After stent placement, the exchange wire was passed through the tract and secured at the PCD site and outside the endoscope channel. A large PCD, preferably greater than 28F, was crucial for facilitating the passage of the second slim endoscope or cholangioscope through the cutaneous fistula following PCD removal. A 7 Fr × 7 cm double pigtail stent was positioned over the wire. Two stents were deployed between the gastric lumen and the tract. The first stent was advanced through the working channel of the scope, whereas the second stent was deployed either through the scope or through the external fistula site. The use of the two stents was intended to mitigate the risk of migration, ensure tract patency, and maintain internal drainage during tract maturation. This double-scope approach is not mandatory. It is only used in selected complex cases where intraluminal visualization and accessory controls are critical. Here, another slim endoscope or cholangioscope provides direct visualization of the fistulous tract and guidewire, enhancing technical precision, reducing procedural risk, and ensuring accurate stent deployment ([Fig. 7]).

• B. Interventional radiologist-guided rendezvous techniques

  This technique can be performed both outside-to-inside and inside-to-outside, in a rendezvous manner. It involves both an experienced endoscopist and a radiologist. In the outside-to-inside rendezvous technique, the existing PCD was used to puncture the stomach and duodenum using a transjugular intrahepatic portosystemic shunt needle; thereafter, the needle guidewire was manipulated into the stomach and captured endoscopically. Subsequently, the transmural tract was dilated, and the stents were placed. In the inside-to-outside rendezvous technique, EUS-guided needle puncture of the fistula tract was done. Thereafter, the guidewire was grasped by the interventional radiologist (IR). The tract was dilated over the guidewire, and a transmural stent was placed. Double scope technique describes purely EUS-based inside-to-outside puncture with complete endoscopic control. In contrast, interventional radiologist guided rendezvous technique represents a hybrid IR-endoscopy rendezvous hybrid technique where the wire introduced endoscopically is retrieved externally by the IR team to facilitate secured tract creation, external manipulation, and combined internal–external drainage. This collaboration is especially useful in long or angulated tracts or when the guidewire cannot be advanced entirely endoscopically. It is one of the ways of doing tractogastrostomy as described in the study by Irani et al[5] ([Fig. 4]).

Fig. 7 Artificial fluid collection technique. Red arrow: 19G needle. Yellow arrow: Fluid collection.

Endoscopic Ultrasound-Guided Pancreatogastrostomy

In patients in whom all the above techniques failed (failure to localize the fistulous tract, failure to create an artificial pancreatic collection, the tip of the pigtail catheter was not adjacent to the gastroduodenal lumen, and the diameter of the PD of the disconnected pancreas was >4 mm), direct puncture of the PD was performed. The dilated PD was localized in the upstream disconnected pancreas and punctured using a 19G needle. The guidewire was then pushed and coiled until the tail end of the pancreas was disconnected. The transmural tract was dilated with a 6 Fr cystotome followed by the placement of a 5 Fr plastic stent ([Fig. 8] and [9]).

Fig. 8 Stent position seen endoscopically.

Fig. 9 EUS-guided pancreatogastrostomy. Red arrow: 19G needle. Blue arrow: Dilated pancreatic duct. Purple arrow: PCD. EUS, endoscopic ultrasound; PCD, percutaneous catheter drainage.

Technique Selection

Practical Selection Guidance (How to Choose)

• If the PCD tip/tract is clearly visualized on EUS: Start with direct puncture or wire localization method, these are efficient and low complexity methods.

• If the tract is not visible but communicates with PCD: Try water instillation or artificial fluid creation method.

• When there is a dilated upstream PD and other method fails: EUS-PG offers a duct-directed risk but requires expertise and carries a higher technical risk.

• Direct PCD puncture method: Useful when the catheter and its exit are readily accessible and confirmable endoscopically.

• For complex anatomy, angulated long tracts, or failed endoscopic attempts: Double scope methods or consider the IR-guided rendezvous method.

Challenges in Routine Practice

Anatomical and Visualization Challenges

• Collapsed or non-distended collections and tracts are difficult to identify and target safely.

• Intervening vessels (small collaterals, perigastric varices) increase the risk of bleeding.

• Long distance from the gastric wall: Tract creation is difficult and unstable.

• Obesity or altered anatomy restricts adequate visualization and EUS maneuverability.

Guidewire and Access-Related Challenges

• Loss of guidewire access during exchange or dilation is common due to tough fibrotic tract walls.

• Kinking of wire inside necrotic debris.

• Wire not coiling inside artificially created collections leads to false tract or extraluminal placement.

• In the PCD-assisted technique: Difficulty identifying the PCD wire under EUS due to echogenic clutter.

Tract Dilation Challenges

• Dense fibrosed tract is associated with poor compliance and requires multiple dilations.

• Risk of perforation: When dilating an immature tract.

• Bleeding: Small vessel injury during dilation may lead to bleeding.

• Cystotome not advancing when there is inadequate apposition between the tract and the gastric wall.

Stent Deployment Challenges

• Maldeployment of the stent

• Stent migration

• Stent clogging by thick debris

Infection-Related Challenges

• Fever or sepsis after procedure requiring antibiotics due to inadequate drainage despite stent placement.

• Secondary infections

Patient-Related Challenges

• Coagulopathy

• Poor tolerance of sedation in critically ill patients

• High inflammatory states: Friable tissue and delayed healing.

Technical Expertise and Challenges

• Non-tertiary centers lack adequate expertise and skilled EUS operators comfortable with access techniques.

• Fluoroscopy: Not readily available across all centers.

• Team coordination (IR, anesthesia): Not readily available across all centers.

Postprocedure Challenges

• Stent-related complications: Bleeding, migration, infection.

• Recurrence or incomplete resolution: Need for reintervention.

• Difficult to determine optimal stent removal timing in DPDS.

Documentation and Follow-Up Challenges

• Poor patient follow-up may lead to recurrence or chronic fistula.

• Centers lacking standardized protocols for follow-up imaging, stent removal timing, approach to DPDS diagnosis and management.

Predictors of Success/Failure

• A distance of <1 to 1.5 cm between the gastric wall and the tract predicts success, whereas >1 to 1.5 cm favors failure.

• Intervening blood vessels favor failure of the procedure, whereas the lack of blood vessels favors success.

• Previous interventions and tract fibrosis causes reduce tract compliance and favor failure.

• High-volume endoscopists demonstrate better outcomes, fewer complications, and improved wire control.

• Preprocedure antibiotics and absence of septic instability improve procedural safety and resolution, and favors success of the procedure.

Rescue Techniques Used in Cases of Persistent Drain Output Even after Tractogastrostomy

Close Proximity of Percutaneous Catheter Drainage and Plastic Stent

If the PCD and plastic stents are placed in close proximity, owing to the larger diameter of the PCD and the higher pressure gradient compared with the plastic stent, the majority of pancreatic fluid drains into the PCD. In such cases, the position of the PCD was adjusted by a radiologist, and the PCD was pulled out to allow pancreatic fluid to drain through the plastic stent into the stomach.

Stent Migration

Persistent drain output even after tractogastrostomy may occur in case of stent migration, which requires restenting. The stent migration rate in tractogastrostomy was approximately 15% to 30% in previous studies.[6] [7] In our experience, stent migration rates were 14.28%; however, long-term data and follow-up are still required to assess migration rates. All three patients who had undergone stent migration in our study had cavity resolution at the time of migration, and the fistula had matured; no reintervention was required.

Stent Upsizing

In a few patients with persistent drain output even after the procedure, stent upsizing was required to create a chronic internal fistula.

Results

A total of 21 patients with EPF secondary to DPDS underwent EUS-guided tractogastrostomy using various technical approaches. [Table 2] provides a side-by-side comparison of the various techniques used for internalization. The techniques employed included direct tract puncture (n = 4), water instillation (n = 2), artificial PFC creation (n = 3), wire localization (n = 3), and the double-scope technique (n = 2). Tract localization was further aided in some cases by direct PCD puncture (n = 3), EUS-PG (n = 2), and IR-guided access (n = 2). All patients had a prior ERCP, with a mean of one PCD drain in situ at the time of the procedure. The average daily drain output ranged from 150 to 250 mL, with the highest output observed in the wire localization and IR-guided groups (250 mL each). Drain fluid amylase levels were elevated in all cases, with a mean of 6,250 to 16,000 U/L. The mean procedural duration for all tractogastrostomy procedures combined was 54.3 minutes. The shortest procedural time was observed with direct EUS-guided puncture techniques, with a minimum duration of 30 minutes with the wire localization technique, whereas the longest procedural time was noted in hybrid techniques (double-scope or IR-guided rendezvous), extending up to 60 to 90 minutes, reflecting increased technical complexity and the need for multidisciplinary coordination. Minor adverse events were observed in selected patients, including mild postprocedural abdominal pain (lasting 3–5 days) in four patients and minor bleeding in one patient using direct tract puncture. No major complications were noted. At 1-week postprocedure ultrasound, no residual fluid collections were observed in any patient. At the 3-month follow-up, complete closure of the EPF was achieved in all 21 patients (100%). The mean follow-up duration was 6 months, during which time no recurrence of fistula or need for repeat intervention was noted. Three (14.28%) patients experienced stent migration during 3-month follow-up X-ray imaging (one each in the direct tract puncture, double-scope, and IR-guided technique groups), although no additional intervention was required. Migration was predominantly asymptomatic and identified during scheduled follow-up imaging or endoscopy. No migration-related perforation or major adverse event was recorded. In all three cases, migration occurred after fistula resolution and cavity collapse; therefore, no further intervention was needed. The tractogastrostomy had already matured, allowing spontaneous internal drainage until healing. No cases of delayed bleeding or infection were reported during the follow-up.

Table 1 Procedural techniques for tractogastrostomy: Indications, pros, and cons

Category	Technique	Primary Indication	Pros	Cons/Limitations
EUS-guided direct puncture techniques	Direct puncture of fistulous tract	Fistulous tract or PCD tip clearly visualized on EUS	• Simple, short, and direct procedure • Uses standard EUS accessories • Minimal instrumentation and lower procedural complexity	• Requires excellent EUS visualization • Risk of misplacement if tract poorly defined • Potential leakage or inadequate drainage
	Wire localization technique	Tract not clearly visible on EUS, but PCD in situ	• Improves alignment and precision • Useful in angulated or tortuous tracts	• Additional technical step • Initial wire is only a reference • Risk of wire kinking/dislodgement
	Direct PCD puncture technique	PCD is clearly visualized and adjacent to the gastric wall	• Highly specific targeting • Immediate confirmation with dye • Can shorten procedure	• Risk of puncturing necrotic tissue/vessels • May not yield durable internalization • Increased complexity/accessory use
Water instillation techniques	Air bubble localization technique	Tract and PD not visualized on EUS	• Enables localization of otherwise invisible tract • Simple adjunct using existing PCD	• Transient visualization • Risk of false localization • Ineffective in loculated/non-communicating tracts
	Artificial fluid collection creation technique	No visible tract or PD; other localization methods failed	• Converts non-visible anatomy into a target • Avoids new punctures	• Requires a 24- to 48-hour delay • Risk of infection/collection expansion • May fail if collection does not enlarge
Hybrid techniques	Double-scope technique	Complex anatomy, difficult wire control, failed simple methods	• Direct visualization from both ends • Superior precision in complex tracts • Dual stents reduce migration	• Resource- and expertise-intensive • Requires large-bore PCD (>28F) • Longer procedure time
	IR-guided rendezvous technique	Long, angulated, or difficult tracts; failed endoscopic access	• High success in challenging anatomy • Secure tract creation • Enables hybrid internal–external drainage	• Logistically complex • Requires IR expertise and equipment • Radiation exposure and infection risk
Alternative drainage strategy	EUS-guided pancreatogastrostomy	Failed tract localization; dilated upstream PD (>4 mm)	• Definitive internal drainage • Independent of fistula/PCD anatomy	• Technically demanding • Risk of pancreatitis/bleeding • Requires long-term stent surveillance

Abbreviations: EUS, endoscopic ultrasound; EUS-PG, EUS-guided pancreatogastrostomy; PCD, percutaneous catheter drainage.

Table 2 Summary and clinical outcomes of the patients

Techniques	EUS-guided direct puncture techniques				Water instillation technique		Hybrid techniques		Alternative drainage strategy
	Direct tract puncture		Wire localization technique	Direct PCD puncture	Air bubble localization technique	Artificial PFC technique	Double scope technique	Interventional radiologist-guided technique	EUS-PG
Number	4		3	3	2	3	2	2	2
Prior ERCP	Yes		Yes	Yes	Yes	Yes	Yes	Yes	Yes
Number of PCD drains (av.)	1		1	1	1	1	1	1	1
Daily drain output (av.)	200 mL		250 mL	250 mL	180 mL	150 mL	150 mL	200 mL	150 mL
Drain fluid amylase (av.)	12,000		11,030	7,300	6,250	9,900	16,000	11,000	8,300
Procedure duration (minutes)	45		30	45	60	60	60	90	45
Postprocedure complications	Minor bleeding		Nil	Abdominal pain (3–5 days)	Nil	Nil	Abdominal pain (3–5 days)	Abdominal pain (3–5 days)	Abdominal pain (3–5 days)
1-week USG	No collection		No collection	No collection	No collection	No collection	No collection	No collection	No collection
1-month USG
3-month USG
EPF closure at 3 m	Yes		Yes	Yes	Yes	Yes	Yes	Yes	Yes
F/U duration (av.)	6 m		6 m	6 m	6 m	6 m	6 m	6 m	6 m
F/U complications (number)	Stent migration (on 3-month X-ray imaging)	1	Nil	Nil	Nil	Nil	1	Nil	1
	Bleeding	Nil	Nil	Nil	Nil	Nil	Nil	Nil	Nil

Abbreviations: EPF, external pancreatic fistula; ERCP, endoscopic retrograde cholangiopancreatography; EUS, endoscopic ultrasound; EUS-PG, EUS-guided pancreatogastrostomy; PCD, percutaneous catheter drainage; PFC, pancreatic fluid collection.

Discussion

The prevalence of DPDS has been reported to range from 30% to 50% in previous studies. It is almost exclusively associated with ANP and not interstitial pancreatitis.[8] [9] [10] [11] In a retrospective analysis, the most common site was the pancreatic neck (57.8%), followed by the distal body tail (23.1%) and the mid body (19.2%).[12] Clinical manifestations can be highly variable and depend on multiple factors, including the underlying etiology, site of disruption (proximal/distal, anterior/posterior), extent of disruption (complete/partial), and the total volume of pancreatic secretion. A small leak from the side branch, especially one associated with an unobstructed MPD, usually heals spontaneously. Large leakages from the MPD, especially if associated with downstream obstruction, require endoscopic or surgical intervention. Persistent leakage of pancreatic fluid into nearby structures and spaces leads to the development of PFC and fistulae.[13] Pancreatic fistula refers to an unusual connection between the epithelium of the PD and other epithelial surfaces. It may be anatomically classified based on the site of involvement of internal or external fistulas. In patients with an EPF, the drain output should be collected and sent for fluid amylase measurement.[14] Generally, an amylase level three times the normal serum level (usually >1,000 IU/L) is suggestive of pancreatic fistula.[15] Computed tomography (CT) is a valuable imaging tool for assessing pancreatic parenchyma and detecting and locating fluid collections. ERCP is considered the gold standard for diagnosing duct disruption by directly visualizing ductal anatomy following contrast injection into the main PD via the papilla. MRCP assesses the pancreatic parenchyma and outlines the ductal structure without cannulation of the papilla, similar to ERCP.[16] Potential treatment strategies include conservative management, endoscopic drainage (transmural/transpapillary), surgical drainage, and combination therapies. Supportive care, including antibiotic and nutritional therapy, is crucial for patients with DPDS. However, conservative treatment alone is unlikely to result in resolution. [Figure 10] shows the algorithm clinical pathway for patients with DPDS.

Fig. 10 Algorithm clinical pathway for patients with DPDS. DPDS, disconnected pancreatic duct syndrome; EUS-PG, endoscopic ultrasound-guided pancreatogastrostomy; IR, interventional radiologist; MRCP, magnetic resonance cholangiopancreatography; PCD, percutaneous catheter drainage; PFC, pancreatic fluid collection.

EUS-guided tractogastrostomy is a relatively novel technique, and the existing literature is limited. [Table 3] provides a comparative analysis of various techniques used in previous studies. Initial work by Arvanitakis et al demonstrated 100% success with transmural drainage and EUS-guided pancreatogastrostomy (EUS-PG). Follow-up in the study was performed for up to 18 months; three patients had fistula recurrence.[17] The study by Will et al demonstrated 100% success with EUS-guided transluminal gastropancreaticostomy with no recurrence on follow-up up to 15 months.[18] In a study by Irani et al, 13 of 15 patients underwent either IR-guided transgastric/transduodenal or EUS-guided drainage. A 100% success rate was achieved, with no recurrence for up to 25 months of follow-up. Although complications occurred in 27% of patients, they were mostly self-limited or related to stent migration, highlighting the importance of stent choice and follow-up.[5] Rana et al demonstrated a 94% success rate using various innovative techniques like artificial PFC creation and direct tract puncture, with resolution between 5 and 21 days. However, a notable stent migration rate (29.4%) underscores procedural limitations.[19] In a study by Hayat et al, EUS-guided pancreaticogastrostomy showed sustained clinical and technical success when conventional endoscopic PD drainage failed.[20] Singla et al in four patients combined fluoroscopic and endoscopic techniques to enhance access and control, with favorable mid-term outcomes.[3] More recent efforts by Kadkhodayan and Irani also demonstrated clinical success without major complications, supporting the safety and efficacy of EUS-guided pancreatic rendezvous as a rescue strategy for patients who failed traditional percutaneous and endoscopic therapy.[21] In our study, we evaluated 21 patients and achieved 100% success with no recurrence at 6 months of follow-up. The rate of stent migration was 14.28%. Our study had a few limitations as well: Retrospective, small sample size, lack of a control arm, and heterogeneity of techniques.

Table 3 Comparative analysis of techniques in previous studies

Study	Patients	Intervention	Prior therapy for DPDS	Succes rate	Time until success	Procedure-related complication	Follow-up
Arvanitakis et al (2007)[17]	13	Transmural drainage between the fistula path and gastrointestinal tract (N = 5) and endoscopic ultrasound-guided pancreaticoduodenostomy were performed because of complete pancreatic duct rupture (N = 1)	NR	100%	NR	None	14 months
Will et al (2011)[18]	1	EUS-guided transluminal gastropancreaticostomy	Yes	100%	Postprocedure day 12	Nil	15 months
Irani et al (2012)[5]	15	Outside-in interventional radiologist guided transgastric/-duodenal puncture (n = 10); Inside-out: endoscopic ultrasound-guided fistula puncture (n = 3); reconnecting the disconnected duct: Interventional radiologist-guided transpapillary access and internalization (n = 2)	NR	100%, no recurrences	Median 7 days (range 1–73 days)	No mortality; adverse events in 4/15 (27%): One fever requiring oral antibiotics, five new PFC (two asymptomatic): Four spontaneous migration of transluminal stents, of which three developed a new symptomatic PFC	25 months (range 6–113)
Rana et al (2019)[19]	18	Transmural placement of pigtail stent through gastric opening of trans-gastric PCD (n = 5), EUS-guided transmural puncture of fluid collection created by clamping PCD (n = 5) or by instillation of water through PCD (n = 3), direct EUS-guided puncture of fistula tract (n = 1), and EUS-guided pancreaticogastrostomy (n = 4)	Yes	94%	5–21 days	Stent migration (29.4%)	16.7 ± 12.8 weeks
Hayat et al (2020)[20]	8	EUS-guided pancreaticogastrostomy	Yes	88%	NR	Nil	9 months
Singla et al (2022)[6]	4	EUS-guided rendezvous and tractogastrostomy	Yes	100%	NR	Stent migration (2 patients)	10.5 months (range: 8–12)
Kadkhodayan and Irani (2025)[21]	3	EUS-guided pancreatic rendezvous	Yes	100%	Postprocedure day 20	Nil	33 months

Abbreviations: DPDS, disconnected pancreatic duct syndrome; EUS, endoscopic ultrasound; PCD, percutaneous catheter drainage; PFC, pancreatic fluid collection.

In conclusion, EPFs in cases of DPDS in ANP are associated with high morbidity. EUS-guided tractogastrostomy is a novel, minimally invasive procedure that restores internal drainage by addressing the underlying pathophysiology of DPDS, that is, persistent upstream ductal pressure. However, further prospective studies with larger cohorts and long-term follow-up are necessary to validate the efficacy, refine patient selection, and standardize the procedural protocols.

Conflict of Interest

None declared.

Ethical Approval

The study was approved by the Institutional Ethics Committee of Medanta – The Medicity, Gurugram, Haryana, India (1096, December 31, 2025).

Authors' Contributions

S.S., G.K., and R.S. conceived the study. Data curation was performed by S.S., G.K., and S.G. Formal analysis, investigation, and methodology were carried out by G.K., S.K., and S.G. Resources, supervision, and validation were undertaken by A.S.P., S.S. S.R.M. G.K., and S.K. drafted the manuscript and performed review and editing.

• References

• 1 Alsaad AA, Dhannoon SM, Phillips MB. Disconnected pancreatic duct syndrome: A multidisciplinary management dilemma. BMJ Case Rep 2016; 2016: bcr2016217099
• 2 Sikora SS, Khare R, Srikanth G, Kumar A, Saxena R, Kapoor VK. External pancreatic fistula as a sequel to management of acute severe necrotizing pancreatitis. Dig Surg 2005; 22 (06) 446-451 , discussion 452
• 3 Singla V, Gupta PK, Singh P. et al. Novel treatment with double scope technique for disconnected pancreatic duct syndrome with external pancreatic fistula. Endosc Int Open 2024; 12 (04) E593-E597
• 4 Bang JY, Navaneethan U, Hasan MK, Hawes RH, Varadarajulu S. EUS correlates of disconnected pancreatic duct syndrome in walled-off necrosis. Endosc Int Open 2016; 4 (08) E883-E889
• 5 Irani S, Gluck M, Ross A. et al. Resolving external pancreatic fistulas in patients with disconnected pancreatic duct syndrome: Using rendezvous techniques to avoid surgery (with video). Gastrointest Endosc 2012; 76 (03) 586-93.e1 , 3
• 6 Singla V, Arora A, Rana SS. et al. EUS-guided rendezvous and tractogastrostomy: A novel technique for disconnected pancreatic duct syndrome with external pancreatic fistula. J Dig Endosc 2022; 13 (03) 136-141
• 7 Kawaguchi Y, Lin JC, Kawashima Y. et al. Risk factors for migration, fracture, and dislocation of pancreatic stents. Gastroenterol Res Pract 2015; 2015: 365457
• 8 Bang JY, Wilcox CM, Navaneethan U. et al. Impact of disconnected pancreatic duct syndrome on the endoscopic management of pancreatic fluid collections. Ann Surg 2018; 267 (03) 561-568
• 9 Jang JW, Kim MH, Oh D. et al. Factors and outcomes associated with pancreatic duct disruption in patients with acute necrotizing pancreatitis. Pancreatology 2016; 16 (06) 958-965
• 10 Varadarajulu S, Wilcox CM. Endoscopic placement of permanent indwelling transmural stents in disconnected pancreatic duct syndrome: Does benefit outweigh the risks?. Gastrointest Endosc 2011; 74 (06) 1408-1412
• 11 Dua MM, Jensen CW, Friedland S. et al. Isolated pancreatic tail remnants after transgastric necrosectomy can be observed. J Surg Res 2018; 231: 109-115
• 12 Tann M, Maglinte D, Howard TJ. et al. Disconnected pancreatic duct syndrome: Imaging findings and therapeutic implications in 26 surgically corrected patients. J Comput Assist Tomogr 2003; 27 (04) 577-582
• 13 Larsen M, Kozarek R. Management of pancreatic ductal leaks and fistulae. J Gastroenterol Hepatol 2014; 29 (07) 1360-1370
• 14 Lawrence C, Howell DA, Stefan AM. et al. Disconnected pancreatic tail syndrome: Potential for endoscopic therapy and results of long-term follow-up. Gastrointest Endosc 2008; 67 (04) 673-679
• 15 O'Toole D, Vullierme MP, Ponsot P. et al. Diagnosis and management of pancreatic fistulae resulting in pancreatic ascites or pleural effusions in the era of helical CT and magnetic resonance imaging. Gastroenterol Clin Biol 2007; 31 (8-9 Pt 1): 686-693
• 16 Busireddy KK, AlObaidy M, Ramalho M. et al. Pancreatitis-imaging approach. World J Gastrointest Pathophysiol 2014; 5 (03) 252-270
• 17 Arvanitakis M, Delhaye M, Bali MA, Matos C, Le Moine O, Devière J. Endoscopic treatment of external pancreatic fistulas: when draining the main pancreatic duct is not enough. Am J Gastroenterol 2007; 102 (03) 516-524
• 18 Will U, Fueldner F, Goldmann B, Mueller AK, Wanzar I, Meyer F. Successful transgastric pancreaticography and endoscopic ultrasound-guided drainage of a disconnected pancreatic tail syndrome. Therap Adv Gastroenterol 2011; 4 (04) 213-218
• 19 Rana SS, Sharma R, Gupta R. Endoscopic treatment of refractory external pancreatic fistulae with disconnected pancreatic duct syndrome. Pancreatology 2019; 19 (04) 608-613
• 20 Hayat U, Freeman ML, Trikudanathan G, Azeem N, Amateau SK, Mallery J. Endoscopic ultrasound-guided pancreatic duct intervention and pancreaticogastrostomy using a novel cross-platform technique with small-caliber devices. Endosc Int Open 2020; 8 (02) E196-E202
• 21 Kadkhodayan K, Irani S. EUS pancreatic rendezvous to rescue failed ERCP for postoperative pancreatic fistula after pancreaticoduodenectomy. VideoGIE 2025; 10 (10) 557-559

Publication History

Article published online:
12 February 2026