Walled-off pancreatic necrosis: a staged multidisciplinary step-up approach

Abstract

Walled-off pancreatic necrosis (WOPN) is a rare complication of pancreatitis. We present the case of a woman in her eighties admitted for diffuse abdominal pain. She had a palpable abdominal mass and the CT scan showed necrosis throughout the tail of the pancreas, a peripancreatic and retrogastric hydroaerial collection (19 cm of diameter) and a calculus in the main biliary duct, thus establishing a diagnosis of emphysematous necrotising obstructive pancreatitis. A step-up approach was decided, first with removal of the biliary calculus, followed by a waiting period of 4 weeks in which the patient was under intravenous antibiotics. At re-evaluation, the CT scan showed a smaller and more organised collection, bounded by a wall, defining WOPN. At this stage, transgastric drainage via echoendoscopy was attempted, without success, followed by percutaneous CT-guided drainage, also with little effect. Surgical necrosectomy was then executed, as a final step, with a successful outcome.

Background

Walled-off pancreatic necrosis (WOPN) is a severe complication of acute pancreatitis and its treatment must be individualised for each patient. The revised Atlanta classification system defines it as a collection associated with necrotising pancreatitis, that persists after 4 weeks and develops a wall.¹

We present the case of a patient with WOPN secondary to gallstone pancreatitis, which was managed successfully with a multidisciplinary approach involving antibiotic, endoscopic and surgical treatment. We report this case in hopes of contributing to a better understanding of the treatment of this disorder.

Case presentation

A woman in her eighties was admitted for a 1-week history of diffuse abdominal pain, nausea and vomits. She had a history of well-controlled arterial hypertension, medicated with losartan 100 mg/day. On physical examination, she was tachycardic but all other parameters were normal. Her abdomen was distended, with diffuse tenderness to palpation and a solid mass was palpable in the epigastric region and upper left quadrant.

Investigations

Admission blood tests depicted leucocytosis with neutrophilia (19.3×10⁹/L and 18×10⁹/L, respectively), C reactive protein of 23 mg/dL, with a normal bilirubin and amylase value. The abdominal CT scan showed necrosis throughout the tail of the pancreas and an exuberant peripancreatic and retrogastric hydroaerial collection (with 19 cm of greater diameter) that extended to left hypochondrium and right quadrants, as well as a slight prominence of the intrahepatic bile ducts and a distal calculus in the main biliary duct (MBD) and vesicular lithiasis (figure 1). These results allowed us to establish a diagnosis of emphysematous necrotising gallstone pancreatitis.

Figure 1.

CT scan images showing an exuberant peripancreatic and retrogastric hydroaerial collection measuring 19×11×15 cm (red arrows). It extends to left hypochondrium and right quadrants. Panel A: coronal cut with a distal calculus in the main biliary duct (yellow arrow). Panel B: axial cut showing vesicular parietal thickening (white arrow).

Differential diagnosis

Strong radiological evidence of pancreatitis on the CT scan, alongside evidence of calculi in the MBD and gallbladder, made the diagnosis of gallstone pancreatitis clear. Differential diagnosis included any peripancreatic fluid collections, such as pseudocysts and necrotic collections. The CT narrowed the differential to an acute necrotic collection due to its radiological characteristics.

We admitted superimposed infection due to the evidence of an emphysematous (gas-forming) collection, in the absence of digestive fistulas on CT.

Re-evaluation at 4 weeks revealed a more organised collection bounded by a wall, thus establishing the diagnosis of WOPN.

Treatment

Following multidisciplinary reunion involving the general surgery and gastroenterology departments, a two-step intervention was proposed: an endoscopic retrograde cholangiopancreatography (ERCP) for calculus removal, followed by transgastric drainage of the collection via echoendoscopy, the timing of the latter to be determined according to clinical status and the collection characteristics on CT scans.

An ERCP was performed on day 1, with sphincterotomy and biliary calculus removal. An upper gastrointestinal endoscopy was performed at the same time to definitely exclude the presence of a digestive fistula (a possible contributing/secondary cause for this collection).

Manipulation of an infect necrotic intra-abdominal immature collection is risky and has a high rate of complications. Thus, since our patient was clinically stable, we first decided to empirically start treatment with piperacillin plus tazobactam while keeping a watchful eye. This was done with the purpose of diminishing the size of the collection and to allow it to develop a more mature inflammatory wall.

We decided to wait 4 weeks, in inpatient treatment, during which the patient was under nasojejunal tube feeding. She remained clinically stable with controlled abdominal pain and sustained apyrexia. At the end of this period, the CT scan showed a slightly smaller and more organised collection bounded by a wall, defining WOPN (figure 2).

Figure 2.

CT scan demonstrating a slightly smaller and more organised peripancreatic and retrogastric collection bounded by wall—walled-off pancreatic necrosis (red arrow).

Transgastric drainage via echoendoscopy was attempted on 31st day of hospitalisation but was not feasible due to the existence of a large gas component (arising from the collection itself) interposing the side of the gastric wall, which impaired the direct visualisation of the target site to create the cystenterostomy.

Two days later, a percutaneous CT-guided drainage was performed, and a drainage catheter was placed, but became rapidly obstructed, having little effect.

On the 40th day, our patient underwent surgery. A retroperitoneal video-guided necrosectomy was performed through a left-side flank incision, during which a large collection with necrotic debris was identified throughout the body and tail of the pancreas. Necrosectomy was executed (figure 3) and a drain was placed in the pancreatic locus (and later gradually externalised).

Figure 3.

Intraoperative picture showing the debrided necrotic pancreas on top of the sterile pads (red arrow).

Extended-spectrum beta-lactamase producing Klebsiella pneumoniae, Proteus mirabilis and Enterococcus faecium were all identified in the purulent exudate drained, for which the patient fulfilled 21 days of meropenem and vancomycin (according to antibiotic sensitivity test).

Due to prolonged anorexia, the nutrition department was involved. Parenteral nutrition was necessary in the immediate postoperative period, followed by enteral feeding until our patient was able to have an adequate caloric oral intake. Surgical wound care and dressing was executed as indicated by general surgery, with good healing.

Due to the lengthy hospitalisation and high burden of disease, there was a loss of motor function for which the patient started a physiotherapy rehabilitation programme, with good results.

Outcome and follow-up

Ten days after surgery, a follow-up CT scan showed a collapse of the collection, without evidence of other collections or ‘new’ elements in comparison to the previous studies (figure 4).

Figure 4.

CT scan images showing the postoperative collapse of the peripancreatic collection (stars).

She was discharged from the internal medicine and general surgery departments on the 65th day, and stayed in the rehabilitation unit for two more months. She was discharged from the hospital on the 126th day of hospitalisation. At 6-month follow-up, she was doing well.

Discussion

A multidisciplinary step-up approach is currently recommended as the best methodology for the treatment of WOPN. This involves percutaneous, endoscopic, laparoscopic and/or surgical procedures. Even though most institutions use a single drainage technique for the treatment of pancreatic fluid collections, in some cases a combination of these techniques is necessary.²

On diagnosis, it is important to determine the characteristics of the wall surrounding the collection. If the wall is immature, time should be given (usually 4 weeks) in order to obtain a more organised and well-defined inflammatory wall before attempting any drainage.³ Interventions performed in acutely ill patients are associated with a higher risk of septic shock and haemorrhagic complications, therefore increasing morbidity and mortality.⁴ If possible, an initial watchful waiting attitude should be considered. The timing of the drainage should be ultimately determined according to the clinical status of the patient.

Percutaneous drainage can be a definitive treatment and also a bridging therapy before endoscopic or surgical interventions. It is also effective for symptoms control such as abdominal pain and intestinal obstruction due to mass effect.⁴

Multicentric studies showed that endoscopic necrosectomy has been associated with a reduced mortality rate compared with surgical necrosectomy.^{5 6} Nonetheless, endoscopic intervention has some limitations in regards to the location of the WOPN, since those that are further away from the gastric area and located near paracolic and pelvic gutters are not feasible for endoscopic treatment.^{7 8}

Despite the development of multiple non-surgical minimally invasive techniques, surgical necrosectomy is still a standard of reference in the treatment of WOPN.³ New and less-invasive surgical techniques are gaining acceptance over the traditional open necrosectomy via laparotomy, due to better survival rates.² Video-assisted retroperitoneal debridement (VARD) is one of those techniques and has demonstrated less complications, morbidity and mortality than the traditional approach.⁹

In our patient, a proper wall developed around the collection after 4 weeks. The initial drainage procedures were done as recommended by literature, but ultimately the team decided to do a minimally invasive VARD, which was performed with a successful outcome.

Patients with WOPN must be treated in specialised centres with experience in pancreatic surgery and endotherapy.⁸

Learning points.

• Walled-off pancreatic necrosis (WOPN) is a rare and severe complication of pancreatitis.

• The treatment of WOPN must be individualised to the patient and requires a multidisciplinary team.

• Minimally invasive interventions are currently recommended as the first option and a step-up approach is supported by literature.