EUS-Guided Pancreatic Pseudocyst Drainage: Review and Experience at Harbor-UCLA Medical Center

Mehrdad Vosoghi, MD; Shahid Sial, MD; Benedict Garrett, MD; Jack Feng, MD; Tonny Lee, MD; Bruce E. Stabile, MD; Viktor E. Eysselein, MD

In This Article

Endoscopic Drainage

Endoscopic drainage can be performed through the major papilla (transpapillary) or through the gastric (endoscopic cystogastrostomy) and duodenal (endoscopic cystoduodenostomy) walls. Endoscopic drainage can be performed with or without EUS guidance.

Several considerations determine the suitability of a particular patient for endoscopic cyst drainage. These considerations include evaluation for cyst apposition to the stomach or duodenal wall, determination of a site suitable for enterostomy, communication between the cyst and pancreatic duct or biliary duct, and cyst fluid consistency. Solid material in the pseudocyst, which is best seen by endoscopic ultrasound, may preclude complete drainage by the endoscopic approach and may require surgery. In addition, malignancy must be ruled out.

Transpapillary drainage is successful only when the cyst is in communication with the pancreatic ductal system. This communication exists about 60% of the time, and is more common in pseudocysts associated with chronic pancreatitis compared with those associated with acute pancreatitis. After ductal communication is documented with a pancreatogram by endoscopic retrograde cholangiopancreatography (ERCP), a guidewire is passed across the pancreatic sphincter, preferably into the pseudocyst. A pancreatic sphincterotomy is performed to enable better access to the main pancreatic duct for instrumentation. Most groups prefer to pass the stent into the pseudocyst; if this is not technically feasible, the stent is advanced to the site of ductal communication and as close as possible to the pseudocyst. A nasocystic drain can also be placed for additional drainage. A potential serious complication of this procedure is introduction of infection and abscess formation. While it is prudent to administer preprocedure antibiotics to prevent such abscess formation, the best preventive measures are to maintain stent patency and remove the stent as soon as the cyst has been effectively drained. EUS done prior to transpapillary drainage is advisable to exclude liquefied necrosis with debris that may on computed tomography (CT) scan appear as pseudocyst. Transpapillary or transenteric drainage of these liquefied necroses with small-diameter drains may result in a high infection rate and is therefore not advised.[20]

In 1 large series,[21] successful drainage was accomplished in 84% of 117 patients, with a recurrence rate of 9%, a complication rate of 12%, and no deaths. The most frequent complication was acute pancreatitis (5%), followed by secondary pancreatic pseudocyst infection (2.5%) that resolved with stent change. A review of 145 patients by Lo and colleagues[13] showed a technical success rate of 94%, resolution of 85%, recurrence rate of 15%, and a complication rate of 12% (4% pancreatitis, 3% infection, and 5% other complications). One potential limitation of this procedure is rapid stent occlusion. Usually only a small-caliber stent can be placed, and as a result, the consistency of the pancreatic pseudocyst fluid is crucial to success because thick necrotic debris can occlude stents. Even though magnetic resonance imaging (MRI) is an appropriate imaging method to determine whether the pseudocyst contains debris, this can be easily determined by EUS during the examination, thus obviating the need for MRI.

Some clinicians have used nasopancreatic drains placed after pancreatic sphincterotomy, with excellent results.[22]

Whereas pancreatic ductal abnormalities associated with stent treatment of chronic pancreatitis have been described,[23] none have been reported in association with transpapillary pancreatic pseudocyst drainage. In 1 large recent study,[22] downstream duct obstruction was seen on pancreatogram in 15 of 42 patients in whom pancreatic pseudocysts were in communication with the pancreatic duct -- 3 as a result of pancreatic duct stones, and 12 resulting from pancreatic duct strictures. In this study, pancreatic duct stones were removed via balloon or basket and patients with pancreatic duct strictures underwent balloon dilation prior to placement of pancreatic duct stents. The study authors concluded that downstream obstruction frequently complicates pseudocyst communication with the main pancreatic duct, and may result in failure of other drainage options, such as transenteric drainage.

Transenteric drainage includes cystogastrostomy and cystoduodenostomy. Multiple reports have documented a high degree of success using transenteric drainage. Endoscopic cystoduodenostomy is the procedure of choice vs cystogastrostomy, because of increased safety, an improved ability to maintain a perpendicular approach to the cyst during drainage, and the relative dependency of the duodenum compared with the stomach for most pseudocysts. A luminal bulge by the cyst is required. In the absence of a luminal bulge, needle localization could require multiple passes of a finder needle, potentially increasing the risk of complications. For cystenterostomy, the anatomic relationship is paramount. The stomach or the duodenum wall must share a common wall with the pseudocyst. Although this may seem obvious, there are instances in which cyst-like structures may be adjacent to and even indent the stomach, without sharing a common wall.[24] Early on, CT scanning made this anatomic determination, but it is now more accurately made by EUS. Complications for these procedures may include hemorrhage, postprocedural pancreatitis, retroperitoneal perforation, and infection or recurrence due to difficulty maintaining an adequate fistula ( Table 1 ). There is virtually no mortality associated with the procedure. It is important to rule out a downstream pancreatic duct obstruction by ERCP prior to removal of the stents placed into the pseudocysts. These pancreatic strictures may otherwise lead to recurrent cyst formation, as discussed above.

Our literature review of 408 patients in 14 studies showed a success rate of 90%, a bleeding rate of 6%, a perforation rate of 3%, other complications (mostly infection) in 6%, a recurrence rate of 12%, and a 0% mortality rate ( Table 1 ).

Overall, these procedures with low complication, recurrence, and mortality rates appear to be acceptable substitutes for surgical treatment when carried out by experienced endoscopists at well-equipped facilities.

EUS-assisted pancreatic pseudocyst drainage is the latest addition to the armamentarium of the endoscopist for drainage of pancreatic pseudocysts. EUS has been used in the management of pancreatic pseudocyst drainage since first reported by Cremer and colleagues[4] in 1989. Wiersema and associates[25] reported the first case of pancreatic pseudocyst drainage entirely guided by EUS in 1996. In 1998, Vilman and colleagues[26] reported the 1-step EUS-guided drainage of pancreatic pseudocysts using a large-channel endoscopic ultrasound sector scanner, with an electrosurgical needle used for puncture. Inui and coworkers[27] further advanced this 1-step technique by puncturing the pseudocyst without the use of electrocautery. Currently, prototype "hot stents" are under development. Mounted on a needle-knife cutting device, they have the advantage of deploying directly after a transmural incision is made, thus obviating the need for pseudocyst guidewire placement and subsequent stent deployment.

EUS has important applications in aiding pancreatic pseudocyst drainage. These applications include:

  • accurately measuring the distance from the gastrointestinal wall to the pseudocyst (a distance > 1 cm is considered a relative contraindication)[20];

  • identifying gastric varices associated with pancreatitis (endoscopic drainage should not be performed when varices are present);

  • imaging gastric vessels;

  • identifying pseudoaneurysms;

  • identifying debris within pseudocysts that may not be drainable and may increase the risk of infection;

  • differentiating pseudocysts from other cystic lesions of the pancreas;

  • identifying septated cysts; and

  • localizing the puncture site in the absence of a visible luminal bulge.[28]

Thus, this technique has the theoretical advantage of reducing the risk of bleeding, perforation, and, potentially, infection. Although the risk of bleeding is reduced with the use of EUS, it is not completely eliminated. Binmoeller and colleagues[29] found 2 patients who had significant bleeding despite prior EUS.

Our experience in 14 patients, added to that of 11 previous studies, for a total of 99 patients, revealed an EUS-guided pseudocyst drainage success rate of 94%, recurrence rate of 9%, and a complication rate of 1.4%, with a 0 % mortality rate ( Table 2 ).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: