Clinical Pancreatology for Practising Gastroenterologists and Surgeons. Группа авторов

needle and coiled into the pseudocyst, usually under fluoroscopic guidance although the procedure can be performed entirely under EUS control. The tract is then dilated using a balloon dilator up to 10 mm, depending on the thickness of the pseudocyst wall. Dilatation with a graduated (4‐5‐7 Fr) biliary dilator may be required prior to balloon dilatation. The other commonly used approach utilises a 10‐Fr cystotome instead of a 19‐gauge needle. The cystotome device (Cook Medical, Bloomington, IN) comprises a needle knife tip, a distal 5‐Fr inner catheter, and a 10‐Fr outer catheter equipped with a diathermy ring at its distal tip. The pseudocyst is punctured using the needle knife tip and the 5‐Fr inner catheter advanced into the cavity. The 10‐Fr catheter is then advanced over the 5‐Fr inner catheter using the ring diathermy to gain access to the pseudocyst. The inner 5‐Fr catheter is then removed and at this stage two 0.035 inch wires can be introduced, with balloon dilatation performed over one of the wires. Placing two wires facilitates rapid second stent placement (if required) following balloon dilatation as the endoscopic view is often suboptimal due to the rapid egress of fluid following dilatation of the tract. Whichever approach is used, one or more plastic pigtail stents are then placed across the stoma. Whilst many experts favor the placement of two plastic double‐pigtail stents, there is no randomized study showing benefit of multiple stents over one stent. In a retrospective study, there was no treatment benefit of multiple stents over one stent or association with stent size (7 Fr vs. 10 Fr). Large and infected collections may benefit from placement of a nasocystic drain and regular irrigation. Given the significant risk of aspiration, endotracheal intubation for the procedure is recommended, particularly for larger collections that are more likely to decompress rapidly. Additionally, CO₂ insufflation is preferred to air to minimize the risk of air embolism. The utility of multiple stents is to facilitate flow of fluid around and between stents. The stents are generally removed after a minimum of six weeks. There is some evidence that recurrence rate is inversely related to indwelling time. If a disconnected pancreatic duct is identified on secretin‐stimulated MRCP or ERCP, the stents can be left in place indefinitely. Fully covered self‐expanding metal stents can be used. Initially, biliary metal stents were repurposed. More recently, the advent of lumen‐apposing metal stents (LAMS) delivered through an electrocautery‐tipped delivery platform specifically designed for EUS‐guided deployment has greatly simplified the process of endoscopic drainage of pancreatic fluid collections. The placement of LAMS has significantly shortened procedure time, reduced the requirement for skilled assistants, and obviated the need for fluoroscopy. Moreover, at least theoretically, the large‐diameter, lumen‐apposing, wide flange stent design has enhanced not only the drainage efficacy of the device but also overall safety of the procedure, potentially minimizing the risk of perforation and pseudocyst wall dehiscence in the setting of indeterminate adherence to the gastroduodenal wall. However, LAMS have several disadvantages compared to plastic stents, including substantially higher cost, greater risk of delayed bleeding (particularly when left in situ for more than three weeks), and requirement for subsequent exchange for long‐term plastic stents in the subgroup of patients with disconnected pancreatic duct syndrome who would otherwise be at risk of pseudocyst recurrence. There are as yet no randomized controlled trials comparing LAMS and plastic stents in the management of pseudocysts. Available data is conflicting in terms of overall clinical success and adverse event rates [11–13]. However, plastic stents do appear to be more cost‐effective whilst LAMS have been associated with a higher risk of bleeding [12] (Figure 17.2).

Figure 17.2 CT scan showing LAMS in situ.

      Source: courtesy of Muhammad F. Dawwas and Kofi W. Oppong.

Transpapillary drainage of pancreatic pseudocysts by means of stent placement in the pancreatic duct is potentially feasible in the subgroup of patients with small to moderate‐sized pseudocysts that clearly communicate with the main pancreatic duct. In this setting, the pancreatic duct is cannulated via the major or minor papilla in the standard fashion and the disrupted segment of the duct is bridged with one, or preferably two, plastic stents after undertaking a pancreatic sphincterotomy. If this is not feasible, placement of the stent tip medial to the leak or within the pseudocyst may also be effective. In practice, however, transpapillary drainage is not recommended as sole drainage therapy for the majority of pancreatic pseudocysts for several reasons. First, given the fact that at least 50% of pancreatic pseudocysts develop in the absence of evidence of disruption of the main pancreatic duct, transpapillary stent placement in this setting would not be anticipated to facilitate effective drainage. Second, undertaking ERCP in the setting of extrinsic luminal compression of the gastroduodenal lumen and gross inflammation of the duodenal mucosa can be technically challenging, as would attempts to stent a disrupted duct lateral to an obstructing stone or high‐grade stricture. Third, preoperative evaluation of the integrity of the main pancreatic duct may not be feasible as a result of unavailability of a high‐quality secretin‐stimulated MRCP service. Last, the smaller diameter of both pancreatic stent and pancreatic duct, compared with the diameter of transluminal stents and fistula, may result in inadequate drainage, increasing the risk of infection in the setting of contrast contamination of an otherwise sterile pseudocyst.

      Combined transluminal and transpapillary drainage is favored by many experts, although the evidence base supporting this practice is less than compelling. If ERCP is performed, the timing of the procedure is also controversial. Undertaking ERCP prior to transluminal pseudocyst drainage may be accompanied by multiple technical challenges as already outlined. On the other hand, delaying ERCP may potentially result in missing a valuable window of opportunity for stenting a potentially bridgeable pancreatic duct disruption that, in the absence of intervention, may evolve into a non‐traversable, high‐grade stricture or even a disconnected duct.

Disconnected Pancreatic Duct Syndrome

      Disconnected pancreatic duct syndrome is a clinical entity characterized by recurrent pancreatitis and persistent extraductal leakage of pancreatic secretions from a disconnected pancreatic tail as a result of complete transection of the pancreatic duct, usually occurring in the setting of a severe episode of acute pancreatitis [14]. The syndrome complicates nearly two‐thirds of pancreatic necrotic collections and one‐third of other pancreatic fluid collections. The diagnosis can be made with varying degrees of accuracy by secretin‐stimulated MRCP, ERCP, EUS, and contrast‐enhanced CT. Successful management requires effective drainage of the disconnected pancreatic segment. If transluminal stent‐assisted drainage of the pseudocyst was undertaken, the most common approach is to maintain the cystenterostomy patent by keeping the transluminal stents in place indefinitely. This is one of the most compelling arguments for recommending the use of plastic over metal stents for transluminal drainage of pancreatic pseudocysts. Transpapillary stent placement into the disconnected duct segment is technically challenging but is not impossible and certainly worth trying. EUS‐guided transluminal placement of a plastic stent into the disconnected duct segment has also been described. Ultimately, if all endoscopic endeavors prove unsuccessful, surgical resection of the disconnected pancreatic segment or even total pancreatectomy with islet cell autotransplantation may be the only therapeutic option.

Complications

      Complications related to endoscopic drainage occur in 5–25% of patients. Transluminal drainage is associated with bleeding, infection, and perforation. Cyst infection occurs independent of the size of the stent used and can even occur when the cavity appears to have completely drained at the time of stent insertion. Periprocedural antibiotics and for a few days after the procedure are therefore advocated, although there is no randomized controlled trial evidence to support their use [3–7]. Repeat imaging is indicated if infection persists or occurs some time after stenting to assess for stent dysfunction which may require further endoscopic intervention.

      Significant bleeding at the time of stent placement is uncommon as the use of EUS guidance should prevent injury to significant