Secondary Logo

Journal Logo

Original Article: Pancreatology

Endoscopic Treatment of Walled-off Necrosis in Children: Clinical Experience and Treatment Outcomes

Bang, Ji Young; Varadarajulu, Shyam

Author Information
Journal of Pediatric Gastroenterology and Nutrition: September 2016 - Volume 63 - Issue 3 - p e31-e35
doi: 10.1097/MPG.0000000000001269


What Is Known

  • Walled-off necrosis is associated with a mortality rate of 10% to 20%.
  • Presently data on the efficacy of endoscopic treatment of WON in children are limited.

What Is New

  • Incorporation of endoscopic treatment as a major part of a multidisciplinary team approach results in successful clinical outcomes in children with walled-off necrosis.

Pancreatic necrosis develops as a consequence of acute pancreatitis in approximately 20% of patients with superimposed infection occurring in about 30% of these cases (1,2). Infected pancreatic necrosis is associated with significant morbidity and mortality rate of 8% to 39% (3). The liquefaction of parenchymal necrosis can mature into a collection termed “walled-off necrosis” (WON), which can be located within the pancreatic parenchyma or peripancreatic space.

Although the majority of WON can be treated conservatively with nutritional and intensive supportive care, a persistent, symptomatic collection that causes pain, gastric outlet or biliary obstruction, persistent or new-onset organ failure, or infection is associated with the mortality rate of 15% to 20% and requires necrosectomy and drainage (4). Unlike uncomplicated pancreatic pseudocysts that can be treated successfully with endoscopic or surgical cystogastrostomy in over 90% of patients, WON is more complex and requires multiple reinterventions and multidisciplinary care (5). Although traditionally treated with open surgical necrosectomy, the management of WON has undergone considerable changes in the past decade. Evidence suggests that minimally invasive surgical techniques are superior to laparotomy with fewer adverse events and lower rates of morbidity and mortality (6). Also, minimally invasive endoscopic techniques such as single or multigate transmural stent placement, sometimes performed in conjunction with nasocystic or percutaneous drain placement and direct or endoscopic percutaneous necrosectomy have gained increasing acceptance with adverse event rates of 10% to 25% (7–11). Although numerous studies have reported on the clinical outcomes of endoscopic treatment in adult patients with WON, the data on its effectiveness in children are limited (7–13). Therefore, the objective of this retrospective study was to evaluate the treatment outcomes of endoscopic management of WON in children.


Patients and Study Design

This is a retrospective study comprising all patients younger than 19 years who underwent endoscopic ultrasound (EUS)-guided drainage of WON over a 6-year period (April 2009–May 2015) at the University of Alabama at Birmingham and Florida Hospital in Orlando. The presence of WON was confirmed by a pancreatic protocol computed tomogram (CT) that was performed during the week before the drainage procedure (14) and was defined as the presence of >30% solid debris within the pancreatic fluid collection. Procedural indications included infection (based on the clinical criteria according to the revised Atlanta classification) (14), gastric or biliary outlet obstruction, persistent abdominal pain or continued systemic illness with failure to thrive despite conservative therapy. The WON was not aspirated for culture before endoscopic drainage, to minimize the risk of secondary infection. Procedural details were entered into a prospectively maintained database. Patients aged 19 years or older or those with irreversible coagulopathy were excluded. Outcomes in 1 of the patients included in the study have been previously reported (13). The study was approved by the institutional review board (IRB approval #754065; X051223006).

Procedure Technique

All procedures were performed by a single experienced endoscopist (SV; approximately 1000 EUS procedures performed per year and 400 pancreatic fluid collection drainage procedures performed to date) under general anesthesia using a linear array therapeutic echoendoscope (GF UCT180T, Olympus America Inc, Center Valley, PA). Intravenous piperacillin/tazobactam or ciprofloxacin were administered to all patients before the procedure and continued for at least 3 days postintervention to prevent secondary infection of the WON. If however, drainage was performed for infected WON, antibiotics were administered for 7 to 10 days. WON was first assessed using EUS to ensure sufficient proximity to the gastric wall (<1.5 cm) and absence of any intervening vasculature. The choice of plastic stents or lumen-apposing metal stents (LAMS) for WON drainage was left to the discretion of the treating endoscopist because in a recent systematic review, no significant difference in treatment outcomes was observed between the 2 stent types (15).

Plastic Stents

WON was punctured using a standard 19G fine needle aspiration needle (Expect needle, Boston Scientific Corporation, Natick, MA), through which a 0.025-in. guide wire was coiled into the necrotic cavity under fluoroscopic guidance. The transmural tract was then sequentially dilated using a 4.5Fr cannula and a 12- to 15-mm radial expansion balloon (CRE balloon, Boston Scientific Corporation, Natick, MA). Two 7Fr, 4-cm double pigtail plastic stents were inserted through the dilated transmural tract into the WON (Fig. 1; Supplemental Digital Content 1, Video 1, Owing to the size (3.7 mm) of the echoendoscope working channel that permits easier deployment of 7 or 8.5Fr stents than the larger diameter 10Fr stents and the technical challenges associated with deployment of 10Fr stents alongside another guidewire, only 7Fr stents were used in the present study. Also, marks were made on the plastic stents at the start of the distal pigtail to inform the endoscopist performing the procedure when it would be safe to start deployment of the stent, and thus minimizing the risk of the stent placement into the WON.

A large WON is seen in the pancreatic tail on EUS (left); the transmural tract is dilated to 12 mm using an over-the-wire radial expansion balloon (center) and 2 7Fr, 4-cm double pigtail plastic stents are placed into the tract for WON drainage (right). EUS = endoscopic ultrasound; WON = walled-off necrosis.


Through-the-scope LAMS measuring 15 mm in diameter with 10-mm saddle length mounted on a 10.8Fr electrocautery-based catheter delivery system were used (AXIOS, Xlumena Inc, Sunnyvale, CA) in the present study. This stent had 2 flanges at either end of the saddle to appose the gastric/enteral wall with the WON in an attempt to minimize the risk of perforation. The WON was accessed using the electrocautery-based catheter delivery system and the distal flange of the LAMS was first deployed under EUS-guidance. This was followed by deployment of the proximal flange within the gastric/enteral lumen under endoscopic view (Fig. 2; Supplemental Digital Content 2, Video 2,

A large WON in the pancreatic body is seen on EUS (left); a transgastric lumen-apposing metal stent is placed for WON drainage (right). EUS = endoscopic ultrasound; WON = walled-off necrosis.

Single transmural tract was created in patients with WON <120 mm in size, with the placement of either 2 7Fr plastic stents or single LAMS. Multiple transmural tracts were created (multiple transluminal gateway technique) in patients with WON at least 120 mm in size and with extension into the flanks (12). The type of prostheses inserted into the additional tracts was of the same type placed into the initial tract created. Antibiotic solution was generally not used for irrigation unless gas was seen within WON on cross-sectional imaging suggestive of an ongoing infection. In such cases, the WON was irrigated using a 120-mg gentamicin admixed with 250-mL normal saline. Direct endoscopic necrosectomy was not performed at the time of the index drainage procedure as the large volume of fluid and debris precludes adequate visualization and hence the ability to perform necrosectomy safely. All patients returned to the hospital wards or the intensive care unit post procedure. Percutaneous gastrojejunostomy tubes (PEG-J) were inserted at the time of the index drainage procedure in patients unable to tolerate oral intake and enteral nutrition via the PEG-J was initiated following the drainage procedure.

72-hours Postprocedure

A CT scan was performed in all patients at 72 hours postdrainage to assess the size of WON, and the patient's response to therapy was reviewed by a multidisciplinary pancreaticobiliary team. In patients without satisfactory radiological and symptomatic improvement following endoscopic drainage, further interventions were performed comprising endoscopic drainage via creation of additional transmural tracts and/or insertion of a large-bore 14Fr percutaneous drain for irrigation and lavage of the necrotic cavity. In patients requiring direct endoscopic necrosectomy, hydrogen peroxide was not instilled into the WON to aid in necrosectomy as this has not been conclusively proven to be beneficial in well-designed trials. Surgical intervention was performed following failure of both endoscopic and percutaneous drainage.

8-week Follow-up

All patients underwent a postprocedure CT scan at 8 weeks to assess treatment response (Supplemental Digital Content 3, Fig. 1, Endoscopic retrograde cholangiopancreatography (ERCP) was performed at this time to examine the pancreatic duct (PD). In patients with WON 2 cm or less and intact main PD, the transmural stents were removed endoscopically. In patients with WON 2 cm or less but with disconnected pancreatic duct syndrome (DPDS), the transmural plastic stents were left in place indefinitely to provide a conduit for drainage of pancreatic secretion from the upstream pancreatic gland and thereby minimizing the chances of pancreatic fluid collection recurrence. In patients with WON 2 cm or less but partial PD disruption, an attempt was made to bridge the leak with a transpapillary PD stent and the transmural stents were removed. In patients without adequate radiological and/or symptomatic resolution, further endoscopic drainage was performed after interdisciplinary consultation. PEG-J tubes were removed at this time in patients with successful treatment outcomes if they were able to tolerate sufficient oral intake.

Outcome Measures and Definitions

The primary outcome measure was rate of treatment success following endoscopic drainage of the WON. The secondary outcome measures were rates of technical success, adverse events, recurrence, and reintervention.

Technical success was defined as successful EUS-guided creation of transmural fistulae and deployment of plastic stents or LAMS for WON drainage. Treatment success was defined as a decrease in the size of WON to 2 cm or less on cross-sectional imaging with concurrent resolution of symptoms at 8-week follow-up. Treatment failure was defined as WON >2 cm with persistence of symptoms despite endoscopic therapy, need for surgical intervention, or death at 8-week follow-up. Recurrence was defined as the occurrence of WON following initial treatment success. Reintervention was defined as the need for repeat endoscopic therapy following initial transmural drainage. Definition of adverse events was based on previously published consensus (16).

Statistical Analysis

Categorical variables were summarized as percentages and proportions, and continuous variables were summarized as means with standard deviation and medians with interquartile range (IQR) and range. The dataset was compiled using Microsoft Excel (Microsoft Corporation, Redmond, WA) and statistical analysis was performed using Stata 13 (StataCorp LP, College Station, TX).


Patient and WON Characteristics

Between April 2009 and May 2015, 6 children consisting of 5 girls (83.3%) and 1 boy (16.7%) with median age of 12 years (IQR 11–17) underwent EUS-guided drainage of WON. The most common etiology of WON was idiopathic acute pancreatitis (50%). All WON were located in the body or tail of the pancreas (5 in body, 1 in tail) with median size of 130 mm (long axis, IQR 80–180 mm). Patient and WON characteristics are summarized in Table 1.

Summary of patient characteristics and WON details

Procedure Details

Using EUS-guidance, 2 7Fr double pigtail plastic stents were inserted through a single transmural tract in 4 patients (66.7%) and 1 LAMS was inserted into a single transmural tract in 1 patient. One 12-year-old girl had a large 220-mm multiloculated WON extending into the flanks that was drained adopting the multigate technique using 6 7Fr double pigtail plastic stents inserted through a total of 3 transmural tracts. This patient also required the placement of a large-bore 14Fr percutaneous drainage catheter for lavage of the necrotic cavity. At follow-up ERCP at 8-weeks, this patient was diagnosed with DPDS and hence 2 plastic stents were left in situ indefinitely to drain the viable upstream portion of the pancreatic duct. PD was intact in the remaining 5 patients on follow-up ERCP and hence did not require placement of a transpapillary PD stent or a permanent indwelling transmural stent. Enteral nutrition via a PEG-J tube was required in 3 of the 6 patients (50%) because of the failure to tolerate oral intake.

Treatment Outcomes

Successful endoscopic creation of transmural tracts and stent deployment was achieved in all patients. Treatment success was also achieved in all patients following 1 intervention in 4 patients (66.7%) and 2 interventions in 2 patients (33.3%). No clinical or procedure-related adverse events were observed. All patients were discharged following WON drainage after a median hospitalization duration of 12 days (IQR 5–33 days). At a median follow-up of 781 days (IQR 336–869 days), all patients were asymptomatic with no WON recurrence, and surgical intervention was not required in any patient. PEG-J tubes were removed in all patients following successful treatment response at 8-week follow-up. The procedure details and treatment outcomes are summarized in Table 2 and Supplemental Digital Content 4, Table 1,

Summary of procedure details and treatment outcomes


The present study demonstrates that endoscopic drainage of WON in children is technically feasible, safe, and highly effective. It precludes the need for more invasive surgical interventions with successful clinical outcomes in the majority of patients.

There are presently limited data on the treatment outcomes of WON in children (13). In a study comprising adult patients that compared minimally invasive stepwise approach to open surgical necrosectomy, nearly one-third of the patients treated by percutaneous drainage in the stepwise treatment arm did not require further interventions (6). Therefore, open surgical necrosectomy is rarely needed when minimally invasive endoscopic and percutaneous treatment options are available. Percutaneous drainage as a stand-alone treatment is rarely undertaken as it predisposes to pancreatic-cutaneous fistula formation and hence is usually performed in conjunction with internal endoscopic drainage. There are, however, certain limitations to the endoscopic treatment approach. The procedure is labor-intensive, time-consuming, and may require multiple reinterventions. Endoscopic necrosectomy, in which the endoscope is advanced into the necrotic cavity to perform debridement, is associated with the risk of perforation and bleeding in 5% and 15% of patients, respectively (17). In a study, we proposed an algorithmic approach in which the endoscopic treatment is tailored according to the specific characteristics of each collection (12); unilocular collections <12 cm in size are treated with stents placed into a single transmural tract and those >12 cm in size are treated using the multigate technique. For necrotic collections extending to the flanks and measuring >12 cm in size, drainage is undertaken using a combination of large-bore percutaneous catheters and internal drainage via single or multigate technique. This strategy yielded a treatment success of >90% compared with only 60% when patients were treated using standard endoscopic techniques (12). Also, this approach minimized the need for endoscopic or percutaneous necrosectomy in a significant proportion of the patients.

We adopted a similar approach for treating children with WON and although our experience is limited, the clinical outcomes as outlined in the present study were superior. None of the children required rescue surgery or even endoscopic necrosectomy. The recent use of LAMS for WON drainage has further simplified the procedural complexity; the single-step technique can be performed under EUS-guidance in 5 to 10 minutes and the large diameter of the stent (15 mm) facilitates better drainage of the necrotic contents. In addition, the creation of multiple tracts in patients with large WON expedites clinical recovery by resulting in faster drainage and very few patients require more invasive endoscopic interventions. Furthermore, although we followed all patients at 8 weeks postprocedure to assess for treatment response, repeating the CT earlier at 4 to 6 weeks could be a reasonable option.

Recurrence of pancreatic fluid collections (PFC) occurs in nearly 25% of patients with WON and DPDS (18). The upstream gland toward the pancreatic tail secretes juice that encapsulates to form a collection that cannot drain via the natural conduit (major duodenal papilla); consequently, it causes abdominal pain and the fluid collection tracts to form a pancreatic-cutaneous fistula (19). To overcome this limitation, we presently leave double pigtail plastic stents in situ indefinitely in patients with DPDS to facilitate the drainage of the upstream gland via the enteral fistulae, thereby lowering PFC recurrence rates. None of the children in the present study developed recurrence of PFC at long-term follow-up.

There are several limitations to the present study. The sample size was small and the study was retrospective in design. It is, however, impractical to perform a randomized trial given the rarity with which WON is encountered in children. Furthermore, all procedures were performed by a single experienced endoscopist with particular expertise in pancreaticobiliary interventions. Nevertheless, we believe that our experience provides a framework for pediatric gastroenterologists and surgeons managing children with WON.

The incorporation of endoscopic interventions, in conjunction with a multidisciplinary treatment approach involving interventional radiologists and minimally invasive surgeons, is important for successful clinical outcomes in children with WON.


1. Banks PA, Freeman ML. Practice guidelines in acute pancreatitis. Am J Gastroenterol 2006; 101:2379–2400.
2. Whitcomb DC. Acute Pancreatitis. N Engl J Med 2006; 354:2142–2150.
3. Neoptolemos JP, Raraty M, Finch M, et al. Acute pancreatitis: the substantial human and financial costs. Gut 1998; 42:886–891.
4. Working Group IAP/APA Acute Pancreatitis Guidelines. IAP/APA evidence-based guidelines for the management of acute pancreatitis. Pancreatology 2013; 13:e1–e15.
5. Varadarajulu S, Bang JY, Sutton BS, et al. Equal efficacy of endoscopic and surgical cystogastrostomy for pancreatic pseudocyst drainage in a randomized trial. Gastroenterology 2013; 145:583–590.
6. van Santvoort HC, Besselink MG, Bakker OJ, et al. A step-up approach or open necrosectomy for necrotizing pancreatitis. N Engl J Med 2010; 362:1491–1502.
7. Varadarajulu S, Phadnis MA, Christein JD, et al. Multiple transluminal gateway technique for EUS-guided drainage of symptomatic walled-off pancreatic necrosis. Gastrointest Endosc 2011; 74:74–80.
8. Gluck M, Ross A, Irani S, et al. Dual modality drainage for symptomatic walled-off pancreatic necrosis reduces length of hospitalization, radiological procedures, and number of endoscopies compared to standard percutaneous drainage. J Gastrointest Surg 2012; 16:248–257.
9. Gardner TB, Coelho-Prabhu N, Gordon SR, et al. Direct endoscopic necrosectomy for the treatment of walled-off pancreatic necrosis: results from a multi-center U.S. series. Gastrointest Endosc 2011; 73:718–726.
10. Bakker OJ, van Santvoort HC, van Brunschot S, et al. Dutch Pancreatitis Study Group. Endoscopic transgastric vs surgical necrosectomy for infected necrotizing pancreatitis: a randomized trial. JAMA 2012; 307:1053–1061.
11. Sarkaria S, Sethi A, Rondon C, et al. Pancreatic necrosectomy using covered esophageal stents: a novel approach. J Clin Gastroenterol 2014; 48:145–152.
12. Bang JY, Holt BA, Hawes RH, et al. Outcomes after implementing a tailored endoscopic step-up approach to walled-off necrosis in acute pancreatitis. Br J Surg 2014; 101:1729–1738.
13. Ramesh J, Bang JY, Trevino J, et al. Endoscopic ultrasound-guided drainage of pancreatic fluid collections in children. J Pediatr Gastroenterol Nutr 2013; 56:30–35.
14. Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis—2012: revision of the Atlanta classification and definitions by international consensus. Gut 2013; 62:102–111.
15. Bang JY, Hawes R, Bartolucci A, et al. Efficacy of metal and plastic stents for transmural drainage of pancreatic fluid collections: a systematic review. Dig Endosc 2015; 27:486–498.
16. Cotton PB, Eisen G, Romagnuolo J, et al. Grading the complexity of endoscopic procedures: results of an ASGE working party. Gastrointest Endosc 2011; 73:868–874.
17. Seifert H, Biermer M, Schmitt W, et al. Transluminal endoscopic necrosectomy after acute pancreatitis: a multicenter study with long-term follow-up (the GEPARD study). Gut 2009; 58:1260–1266.
18. Varadarajulu S, Bang JY, Phadnis MA, et al. Endoscopic transmural drainage of peripancreatic fluid collections: outcomes and predictors of treatment success in 211 consecutive patients. J Gastrointest Surg 2011; 15:2080–2088.
19. Nadkarni NA, Kotwal V, Sarr MG, et al. Disconnected pancreatic duct syndrome: Endoscopic stent or surgeon's knife? Pancreas 2015; 44:16–22.

endoscopic drainage; endoscopic ultrasound; lumen-apposing metal stent; pancreatic fluid collection; plastic stent; walled-off necrosis

Supplemental Digital Content

Copyright 2016 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.