How can I be sure that the patient has pseudocysts and/or fluid collections?

Pancreatic fluid collections and pancreatic pseudocysts (PPs) are common sequelae of pancreatitis. A multidisciplinary group in 2012 revised the Atlanta Classification System for Acute Pancreatitis (AP) and defined the following complications:

  • Acute peripancreatic fluid collection: appearing early in the course of AP, in or around the pancreas, and does not have a well-defined wall. These often resolve spontaneously and do not need drainage.

  • Pancreatic pseudocyst: an encapsulated collection of fluid with a well-defined wall, usually outside the pancreas and without necrosis. These usually occur more than four weeks after the initial episode of pancreatitis.

    Continue Reading

  • Acute necrotic collection and walled-off necrosis: necrotizing pancreatitis can lead to an acute necrotic collection (ANC) with variable amounts of fluid and necrosis without a wall, or as a walled-off necrosis (WON) with a mature capsule and a well-defined wall. These are initially sterile but may become infected (termed ‘infected necrosis’).

Within the first 4 weeks of formation, accumulated peripancreatic fluid is described as pancreatic fluid collections. These early fluid collections typically contain no solid debris and lack a well-defined wall. Most of these collections remain sterile and are spontaneously reabsorbed within several week of AP. Determinants of whether an acute fluid collection resolves or persists to form a PP remains unclear, but infection and necrosis are likely contributors. In patients who develop PPs, the acute collection produces an inflammatory response along the serosal surfaces of adjacent organs, creating a fibrous pseudocapsule. Unlike true cysts, PPs do not have an epithelial lining, thus its name, “pseudocyst.

What causes pseudocysts and where are they located

Although the true prevalence of PPs is unknown because of a high number of undetected cases, it is felt that the incidence of PP mirrors that of AP and CP. Most PPs develop after alcoholic pancreatitis; the second most common cause is gallstone pancreatitis, but any cause of pancreatic injury/trauma can cause pseudocysts to develop. Two mechanisms have been proposed for the formation of PPs in CP: (1) PPs may follow acute exacerbations of chronic disease and (2) obstruction of the pancreatic duct or a branch by a stone or fibrosis can result in cyst formation. Saccular dilation from an obstructed duct causes a true retention cyst, and a number of such cysts coalescing can form a well-circumscribed PP. Rupture of the capsule of these PPs causes fistula formation and pancreatic duct disruption with approximately half of these disruptions occurring in the head of the gland.

Data are inconsistent regarding the incidence of PP and main pancreatic duct communication ranging from 6% to 60%.

A tabular or chart listing of features and signs and symptoms

What is the typical appearance and consistency of pseudocysts?

Most PPs are solitary, but multiple PPs are seen in practice, most commonly in CP. Typical PPs are spheroid in shape but some are multiloculated. Size of pseudocysts vary, ranging from subcentimeter lesions to well over 10 cm. Distinguishing between cystic pancreatic neoplasms (Figure 1) and PPs is important, particularly in the setting of CP, where both of these entities can arise.

Loculation of cyst fluid, thick wall, and extensive septation are suspicious for cystic neoplasms. Magnetic resonance imaging (MRI) and endoscopic ultrasound (EUS) are diagnostic tools that are often able to distinguish these features. Fluid within a PP often contains high levels of pancreatic enzymes, amylase, lipase, and trypsin, but low levels of amylase can be seen if the fistula of a PP seals. Older patient populations and irregular-appearing cysts warrant fluid analysis including cytology, tumor markers (CEA and CA 19-9), and staining for mucin. In a multicenter study, with surgical pathology as a gold standard in 111 cases, a CEA level of 192 ng/mL was found to more accurately distinguish mucinous from nonmucinous pathologies when compared to EUS morphology and cytology. See Figure 1.

How can I confirm the diagnosis?


What other diseases, conditions, or complications should I look for in patients with pseudocysts?

What is the natural history of pseudocysts?

Debate exists regarding the rate of spontaneous resolution of PPs. This debate is due in large part to the lack of consistent data in this setting; many studies in the past failed to distinguish peripancreatic fluid collections from PPs, and/or failed to support the diagnosis by further cross-sectional imaging.

In the setting of AP, ductal disruption that leads to extravasation of enzyme-rich fluid and pseudocyst formation may seal off, causing spontaneous regression of PPs. Larger PPs (>5 cm), pseudocysts associated with CP, PPs arising from severe AP/necrosis, multicystic lesions, and thicker-walled PPs are all less likely to spontaneously resolve.

In a study involving 68 patients with PP who were managed expectantly, resolution occurred in 57% of 24 nonoperative patients with appropriate radiologic follow-up; 38% of patients resolved more than 6 months after initial diagnosis. Other studies reflect the role of size on spontaneous resolution, showing that small PPs (<5 cm in diameter) are more likely to resolve.

What is the right therapy for the patient with pseudocysts?

Which pseudocysts should be treated?

Symptoms potentially warranting PP drainage include intractable abdominal pain, an inability to tolerate oral intake, weight loss, jaundice, and fever. These symptoms may reflect potential complications of PP including infection, rupture, hemorrhage, vascular thrombosis, or obstruction of adjacent structures.

An early observational series suggested that the majority of pseudocysts larger than 6 cm in diameter and persisting longer than 6 weeks result in symptoms and complications. In one early study, the risk of complications was reported to be correlated directly with the length of time the pseudocyst was present: less than 6 weeks, 20%; 7 to 12 weeks, 46%; and 13 weeks or longer, 75%. This created an “urban legend” that the majority of PP would inevitably “go bad” (develop complications) if left untreated for periods longer than 6 weeks.

In a different study of 68 PP patients managed without surgery over a mean of 46 months, 6 (9%) developed early complications including intracystic hemorrhage, perforation, and infection. In 83 nonalcoholic patients without signs of CP, 23% developed complications including pain (12), infection (4), fistula (2), and rupture (1) in the first 6 weeks of observation. Patient selection bias likely impacts the variability of data in these studies, making it difficult to ascertain the true complication risk in PPs. Clearly, however, the potential for serious complications speaks to the need for intervention in selected cases.

What treatment options exist?

The management of an individual PP depends largely on the background and experience of the practitioner, in addition to the available local expertise. No consensus exists as to the “correct” intervention for all cases but what is obvious is that surgical, endoscopic, and radiologic interventions all have their roles, and that optimal management of PPs relies on a multidisciplinary approach.

Endoscopic management

Endoscopic drainage has been applied to both both pancreatic pseudocysts and to walled-off pancreatic necrosis. Patients with chronic pseudocysts treated endoscopically are more likely to resolve than those with acute pseudocysts or necrosis. Endoscopic decompression of PPs aims to create a fistula between a fluid-filled cavity and the lumen of the stomach or duodenum, creating a path of least resistance, allowing fluid to flow from the walled-off PPs to the lumen of the GI tract. Endoscopic access to the PP can be achieved by a transmural or transpapillary approach.

Transmural approach

The increased availability of EUS allow for a survey of the PP and surrounding structures, specifically its relationship to the stomach and vasculature. Although not mandatory, given the success of transmural decompression in the pre-EUS era, EUS is strongly recommended in this setting. Standard technique for endoscopic PP decompression is commonly known as a cyst-gastrostomy (or -duodenostomy).

Briefly, the PP or WON is identified by a bulge into the lumen of the stomach or duodenum (EUS allows identification of PPs that do not create a bulge) and punctured using a needle-knife catheter, causing a gush of fluid and allowing guide-wire access into the pseudocyst. Following balloon dilation to 8 to 10 mm, two 10 French double pigtail biliary stents are placed to drain the cavity, with one end of the stent(s) within the cyst and the other “pigtail” end, in the lumen of the stomach or duodenum.

Pigtail stents are preferred because of the risk of migration with straight stents. The pigtail stents remain in place until the PP has resolved, typically after 1 to 2 months.

Transpapillary approach

This approach is possible when the PP communicates with the pancreatic duct and often used in combination with a transmural approach in the same setting. Endoscopic retrograde cholangiopancreatography (ERCP) or magnetic resonance cholangiopancreatography (MRCP) are used to establish the presence of communication between the PP and pancreatic duct.

With ERCP, guide wire access of the pancreatic duct or the cyst can be performed and stent placement can occur to allow drainage out of the ampulla. If direct passage of the stent into the cyst cavity is not possible, an attempt is made to bridge the cyst communication with the pancreatic duct. This stent remains in place until resolution of the pseudocyst, often 2 to 3 months. Placement of a nasocystic drain to facilitate saline lavage of cysts complicated by necrosis and debris is also an option followed by subsequent replacement by a pancreatic stent.

Communicating pseudocysts to the pancreatic duct have been associated with nonsurgical treatment failures, reflecting the importance of pancreatography prior to any PP intervention through ERCP or MRCP. With control of the fistula through stenting across the ductal disruption or into the cyst itself, endoscopic or percutaneous drainage has a much higher rate of success. If the disrupted duct cannot be bridged, as in “dislocated duct syndrome,” the dislocated segment of pancreas will empty its secretions into a peripancreatic fluid collection. This collection may eventually wall-off to become a true PP. Chronic percutaneous drainage to create a fistula, which ultimately will seal off, or surgery are the two potential therapeutic options in this setting.

Percutaneous drainage

Radiologic-guided percutaneous drainage achieves a similar goal of fistulization with additional invasion of an abdominal wall puncture. No evidence exists to confirm that percutaneous drainage is necessarily more sterile than endoscopic drainage. However, in suspected infected necrosis, percutaneous drainage is often pursued because it allows more expeditious tissue sampling.

Percutaneous drainage may have a role in early necrosis or sepsis as a guide to minimally invasive surgery in patients who fail to respond. Percutaneous drainage has been associated with higher mortality rate, higher incidence of complications, and longer hospitalizations when compared to surgical intervention. However, a selection bias is at work here, as percutaneous decompression is often used preferentially in patients with comorbid conditions that may preclude them from more invasive surgical procedures.

Surgical management

The conventional surgical method involves creating a fistula to drain the PP through marsupialization – that is, creating a pouch or an opening to the back wall of the stomach whose edges are sewn back to prevent retraction or early closure. Anatomic cyst topography may also require the creation of surgical cyst-duodenostomy, Roux-en-Y cyst-jejunostomy, and, in rare cases, external drainage.

To date, no randomized controlled data exist comparing endoscopic management of PPs versus surgery. Limited observational series have shown no differences in treatment success or complications with endoscopic compared to surgical management, although hospital length of stay and mean costs may be higher with a surgical approach.

In 74 patients with PPs following an episode of acute pancreatitis, older age, alcoholic etiology, higher Ransom score of previous pancreatitis, poorer nutritional status, and residual necrosis were predictive of worse outcomes in the setting of any invasive treatment of pseudocysts.

Conservative (medical) management

Medical management is most important in the setting of acute pancreatitis with or without pseudocyst formation, specifically with regard to adequate intravenous fluid hydration.

In patients who develop necrotizing pancreatitis, infected necrotic tissue is of great concern and typically warrants surgical treatment, given the high mortality rate with medical therapy alone. Opinions vary regarding the administration of antibiotics in necrotizing pancreatitis in the absence of confirmed infection. Agents that penetrate pancreatic tissue (imipenem, ciprofloxacin, metronidazole) and decontaminate the gut to prevent translocation are often used. In a randomized-controlled trial, no differences in the rate of infected necrosis, systemic complications, or mortality were seen comparing antibiotic prophylaxis with ciprofloxacin/metronidazole to placebo in patients with severe pancreatitis. In the setting of pancreatic necrosis, prophylactic antibiotics are not recommended due to risks of bacterial resistance and fungal superinfection.

From an interventional perspective, infection is the most common complication of PP drainage, particularly when performed endoscopically, as bacterial colonization is inevitable once communication is established between the lumen of the gastrointestinal tract and the pseudocyst. The American Society of Gastrointestinal Endoscopy guidelines recommend prophylactic antibiotics for transpapillary or transmural drainage of a sterile pancreatic fluid collection (including PP) to prevent cyst infection. A fluoroquinolone administered before the procedure and continued for 3 to 5 days following the procedure is reasonable.

What is the most effective initial therapy?

What factors help in selecting the best management option for the patient with a pancreatic pseudocyst?

See Figure 2.

Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.


A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies


Listing of these, including any guidelines for monitoring side effects.


How can I monitor the patient with pseudocyst?

What imaging follow-up is recommended after treatment?

After definitive treatment of PPs and WON, surveillance imaging is often performed. Ultrasonography, computerized tomography (CT), and MRI have all been used in this setting, but no standardized surveillance schedule exists with regard to timing or imaging modality. If a symptomatic patient does not improve shortly after treatment, repeat imaging is certainly indicated. A small series of patients with larger than 15 mm-sized PPs (treated with cystogastrostomy) experienced life-threatening complications due to incomplete drainage, suggesting that endoscopic drainage of large cyst cavities may not provide adequate dependent drainage.

No guidelines exist regarding the implications of size for endoscopic PP decompression, but ensuring timely drainage of large PPs by regular cross-sectional imaging is important in potentially preventing serious complications.


Acute peripancreatic fluid collections arise within 4 weeks of AP and lack a well-defined wall.

Pseudocysts, are composed of pancreatic juice lined by a wall of inflammatory tissue; acute necrotic collections and walled off necrosis are separate entities that are initially sterile but can become infected.

Alcoholic pancreatitis, followed by gallstone pancreatitis, are the most common causes of PPs.

Distinguishing between PP and cystic neoplasms is important, particularly in a background of CP; CEA is the most reliable tumor marker in distinguishing mucinous from nonmucinous lesions.

Larger PPs, pseudocysts associated with CP, WON arising from severe AP/necrosis are less likely to spontaneously resolve.

Endoscopic, percutaneous, and surgical interventions all have a role in PP and WON management, necessitating the need for a multidisciplinary approach to treatment.

EUS is a valuable tool in pursuing transmural pseudocyst decompression, particularly given its ability to survey the cyst and surrounding anatomy.

Pancreatography is recommended prior to treatment to evaluate for PP-duct communication and potential need for stent placement across a disrupted duct.

Surgical intervention, although most invasive, can provide a definitive treatment and allow for necrosectomy if needed.

Follow-up cross sectional imaging is recommended following treatment, particularly if a large PP was treated and/or the patient’s symptoms do not improve soon after intervention.

What's the evidence?

Banks, PA, Bollen, TL, Dervenis, C. ” Classification of acute pancreatitis—2012: revision of the Atlanta classification and definitions by international consensus”. Gut. vol. 62. 2013. pp. 102

Tenner, S, Baillie, J, Dewitt, J, Swaroop Vege, S. ” ACG Guidelines: Management of Acute Pancreatitis”. Am J Gastroenterol. vol. 108. 2013. pp. 1400-1415.

Bradley, EL. “A clinically based classification system for acute pancreatitis”. Arch Surg. vol. 128. 1993. pp. 586-90.

Baillie, J. “Pancreatic pseudocysts (part I)”. Gastrointest Endosc . vol. 59. 2004. pp. 873-9.

Cannon, JW, Callery, MP, Vollmer, CM. “Diagnosis and management of pancreatic pseudocysts: what is the evidence?”. J Am Coll Surg. vol. 209. 2009. pp. 385-93.

Brugge, WR, Lewandrowski, K, Lee-Lewandrowski, E. “Diagnosis of pancreatic cystic neoplasms: a report of the cooperative pancreatic cyst study”. Gastroenterology. vol. 126. 2004. pp. 1330-6.

Andren-Sandberg, A, Dervenis, C. “Pancreatic pseudocysts in the 21st century (part II): natural history”. JOP. vol. 5. 2004. pp. 64-70.

Baron, TH, Harewood, GC, Morgan, DE, Yates, MR. “Outcome differences after endoscopic drainage of pancreatic necrosis, acute pancreatic pseudocysts, and chronic pancreatic pseudocysts”. Gastrointest Endosc. vol. 56. 2002. pp. 7-17.

Baillie, J. “Pancreatic pseudocysts (part II)”. Gastrointest Endosc. vol. 60. 2004. pp. 105-13.

Johnson, MD, Walsh, RM, Henderson, JM. “Surgical versus nonsurgical management of pancreatic pseudocysts”. J Clin Gastroenterol. vol. 43. 2009. pp. 586-90.

Varadarajulu, S, Lopes, TL, Wilcox, CM. “EUS versus surgical cyst-gastrostomy for management of pancreatic pseudocysts”. Gastrointest Endosc. vol. 68. 2008. pp. 649-55.

Soliani, P, Franzini, C, Ziegler, S. “Pancreatic pseudocysts following acute pancreatitis: risk factors influencing therapeutic outcomes”. JOP. vol. 5. 2004. pp. 338-47.

Isenmann, R, Runzi, M, Kron, M. “Prophylactic antibiotic treatment in patients with predicted severe acute pancreatitis: a placebo-controlled, double-blind trial”. Gastroenterology. vol. 126. 2004. pp. 997-1004.

Forsmark, CE, Baillie, J. “AGA Institute technical review on acute pancreatitis”. Gastroenterology. vol. 132. 2007. pp. 2022-44.

Banerjee, S, Shen, B, Baron, TH. “Antibiotic prophylaxis for GI endoscopy”. Gastrointest Endosc. vol. 67. 2008. pp. 791-8.

**The original authors for this chapter were John Baillie and Wood B. Gibbs. The chapter was revised by Dr. Bruce R. Bacon.