Int J Gastrointest Interv 2023; 12(1): 7-15
Published online January 31, 2023 https://doi.org/10.18528/ijgii220037
Copyright © International Journal of Gastrointestinal Intervention.
Hyung Ku Chon1 and Seong-Hun Kim2,*
1Department of Internal Medicine, Institute of Wonkwang Medical Science, Wonkwang University Hospital, Wonkwang University School of Medicine, Iksan, Korea
2Department of Internal Medicine, Research Institute of Clinical Medicine of Jeonbuk National University-Biomedical Research Institute of Jeonbuk National University Hospital, Jeonju, Korea
Correspondence to:*Department of Internal Medicine, Research Institute of Clinical Medicine of Jeonbuk National University-Biomedical Research Institute of Jeonbuk National University Hospital, 20 Geonji-ro, Deokjin-gu, Jeonju 54907, Korea.
E-mail address: shkimgi@jbnu.ac.kr (S.-H. Kim).
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/4.0) which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Despite many advances in the treatment of acute pancreatitis, the mortality rate in severe cases is as high as 30%. Therefore, it is very important to treat the local complications of acute pancreatitis. Pancreatic fluid collections (PFCs) are among the most important local complications of acute pancreatitis. According to the Atlanta classification of acute pancreatitis in 2012, PFCs are categorized as acute peripancreatic fluid collections, pancreatic pseudocysts, acute necrotic collections, and walled-off necrosis. In principle, PFCs with complications require an intervention such as drainage. With the recent development of endoscopic ultrasound (EUS)-guided interventions, EUS-guided drainage or necrosectomy has emerged as an important standard treatment for PFCs. In this article, we present recent advances in EUS-guided drainage and necrosectomy for PFCs in acute pancreatitis, along with a literature review.
Keywords: Complications, Endosonography, Pancreatic pseudocyst, Pancreatitis, Pancreatitis, acute necrotizing
Despite advances in the treatment of acute pancreatitis, 10%–20% of pancreatitis cases develop into moderate or severe acute necrotizing pancreatitis, and in severe cases, the mortality rate can be as high as 30%. Therefore, early treatment is very important.1–3 Pancreatic fluid collections (PFCs), which account for about 40% of all complications, are the most important local complications affecting the prognosis of acute pancreatitis. According to the Atlanta classification of acute pancreatitis revised by international consensus in 2012, PFCs are categorized as acute peripancreatic fluid collections (APFCs), pancreatic pseudocysts (PPs), acute necrotic collections (ANCs), and walled-off pancreatic necrosis (WOPN), depending on the duration and pattern of the PFC (Fig. 1).2,4 A PP is a localized fluid collection without necrosis in the pancreas that develops over time into a cystic structure surrounded by a well-defined inflammatory wall. A PP is usually observed about 4 weeks after the onset of acute pancreatitis, and it is characterized by the absence of a solid component inside the pancreas. In contrast, WOPN is an encapsulated fluid collection that forms as a late complication related to an ANC. Unlike a PP, a necrotic solid component is observed inside the pancreas. Not all PFCs require drainage and intervention.5 In principle, all infected PFCs should be drained in patients who fail to improve with conservative management alone.6 Based on the guidelines of the International Association of Pancreatology/American Pancreatic Association1, the American Gastroenterological Association, and the American Society of Gastrointestinal Endoscopy7, interventions are recommended for the following conditions: symptomatic PFCs (persistent abdominal, flank, or back pain), PFC-related infections (gas within the collection, presence of debris in a pseudocyst), bleeding, luminal obstruction (gastric or duodenal outlet obstruction related to a local mass effect) (Fig. 2), fistulization, biliary obstruction (Fig. 3), rapid fluid accumulation or pseudocyst growth seen on serial imaging, new onset of symptoms in pseudocysts of any size, and WOPN.8,9 For patients with proven or suspected infected necrotizing pancreatitis, if possible, an invasive intervention should be delayed for ≥ 4 weeks after the initial presentation to allow the collection to become “walled-off.”1,7 Moreover, in cases of symptomatic sterile necrosis > 8 weeks after the onset of acute pancreatitis, drainage is recommended.6,7 However, a multicenter study reported that WOPN was observed in 43% of patients even 3 weeks after the onset of acute pancreatitis, so there was no need to wait 4 weeks for drainage.1,10
Drainage in PFCs includes percutaneous drainage, endoscopic drainage, and surgical drainage. The drainage method should be selected according to the pattern of the PFC and the condition of the patient regarding the advantages or disadvantages of each option.11–13 Traditionally, surgical treatment has been used to treat PFCs, with a 91% to 97% drainage success rate. However, surgical treatment is an invasive procedure with general anesthesia and several studies showed that endoscopic drainage and necrosectomy for WOPN had higher efficacy, shorter length of hospital stay, and lower health care cost compared to surgical drainage.14,15 Therefore, endoscopic drainage is primarily recommended in patients with high surgical risk or a PP or WOPN adjacent to the stomach and duodenum.3,13,16 For drainage of an APFC or ANC—the stage before the formation of a PP or WOPN in a PFC—percutaneous drainage, which uses a catheter with a diameter of 8–23 Fr, may be more useful than other methods. Few studies have compared endoscopic drainage and percutaneous drainage, but percutaneous drainage needs repeated procedures, as well as a longer hospital stay, and it involves a higher risk of side effects (e.g., fistula formation) than endoscopic drainage. Therefore, percutaneous drainage is recommended with combined treatment (e.g., irrigation) or when endoscopic drainage is technically difficult or ineffective.3,11,17,18
A recent guideline recommends endoscopic drainage as the initial standard treatment for PFCs, and the endoscopic drainage approaches include transmural and transpapillary drainage.3 Those methods can be performed alone or in combination depending on the location and size, connectivity and disruption to the pancreatic duct (PD), and the internal properties of the PFCs.3 In general, transmural drainage is considered first, but if leakage of pancreatic fluid due to partial rupture of the PD is observed, transpapillary drainage is recommended.3 There are two methods of transmural drainage: endoscopic transmural drainage (ETD) using a gastroscope or a duodenoscope and EUS-guided transmural drainage (EUS-TD) under endoscopic ultrasound (EUS) guidance. EUS provides real-time ultrasound imaging, so it can reduce the risk of bleeding by checking the intervening vessel during the needle puncture. In addition, EUS can detect PFCs that are not swollen in the intestinal lumen and identify the presence of solid or necrotic components in PFCs. Two randomized controlled trials comparing ETD and EUS-TD found that EUS-TD had a significantly higher success rate.7,19–21 In addition, even when ETD has failed, EUS-TD has been successful. Other studies showed that about 2% to 40% of ETD procedures had complications, such as bleeding, perforation, infection, and stent dysfunction, and the procedure was impossible in 42% to 48% of cases.3,5 Therefore, EUS-TD has been established as a standard endoscopic drainage method when transmural drainage for PFCs is required. We focused our review on the recent advances in EUS-TD and necrosectomy in PFCs related to acute pancreatitis that require an intervention.
Before the procedure, it is necessary to check whether the patient meets the indications for EUS-TD and to check the anatomical structures, such as the formation of a well-encapsulated PFC or the possibility of cyst perforation or bleeding during the procedure through a computed tomography scan or magnetic resonance imaging. For successful EUS-TD, endoscopic access to a proper drainage location should be possible, and the distance between the gut wall and the PFCs should be less than 1 cm.22 The patient takes a left lateral position under conscious sedation in an endoscopy unit with fluoroscopy. The proper drainage location should be determined by a linear array EUS. Then, the PFCs are punctured using a 19-gauge (G) needle while avoiding the intervening vessel under real-time imaging with color Doppler, and the needle stylet is removed. After aspiration of 5–10 cc of fluid, the type of stent is determined by checking the properties of the fluid. A Gram stain and culture of aspirated fluids are performed, as well as cystic fluid amylase, lipase, and carcinoembryonic antigen level tests based on the clinical situation. Then a guidewire (0.025–0.035 inches) is inserted through the needle under fluoroscopic and ultrasonographic guidance, and it is coiled 2–3 times inside the PFC. The needle is then gently removed to prevent the loss of the guidewire while maintaining the endoscopic position. Then, fistula tracts are created along the guidewire on the gut wall and cystic wall using a mechanical dilator such as a 6-Fr Soehendra® dilator or a cautery device such as a 6-Fr cystotome. Usually, the type of the stent to be inserted is determined before the puncture based on the characteristics and size of the PFC, the proportion of solid debris, and the thickness of the matured wall. In general, a metal stent may be unnecessary for additional fistula tract dilation, but a plastic stent should be used to perform fistula tract dilatation with a 6-mm or 8-mm balloon dilation catheter or an 8-mm to 12-mm controlled radial expansion (CRE) balloon. At this time, obliteration of the waist is checked under fluoroscopy. Then, a double pigtail plastic stent (7–10 Fr in diameter) is placed under endoscopic and fluoroscopic guidance. To insert multiple plastic stents, it helps to additionally insert the guidewire after re-cannulation using a catheter, but initially inserting two guidewires in advance using a wide diameter catheter can prevent the guidewire loss from the fistula tract. The technique of additionally inserting a nasocystic drainage catheter for cyst irrigation is the same as those with multiple plastic stents. To insert a metal stent (fully covered self-expandable metal stent [FCSEMS]), there is no need to forcefully dilate the fistula tract before inserting the stent. After deploying the metal stent, the guidewire is placed in the metal stent and the fistula tract is dilated to prevent leakage of cystic fluid. A metal stent with an integrated cautery device has recently been commercialized, making it possible to skip the cumbersome step of dilating the fistula tract to insert a metal stent. When the fistula tract is expanded widely or a metal stent is inserted, a large volume of fluid from the PFC is drained into the gut. At this point, it is important to sufficiently aspirate the drained fluid using endoscopic suction to prevent aspiration of the fluid into the lungs (Fig. 4).
The selection of the puncture needle is one of the most important factors contributing to the success of the procedure. Factors to be considered include the diameter of the needle, the shape of the bevel of the needle tip, and the flexibility of the needle. Although a 22-G needle is sometimes used for EUS-guided drainage of the bile duct, a 19-G needle is usually used for EUS-TD. The differences in the length of bevels and materials (stainless steel, nitinol, and cobalt chromium) of each needle can affect the success rate of the intervention. Each needle has a different bevel angle and length. When pulling the guidewire, care must be taken not to damage the Teflon coating of the guidewire.23 An EchoTip Ultra HD ultrasound access 19-G needle (Cook Endoscopy, Winston-Salem, NC, USA) has an atraumatic needle tip type with a beveled stylet that can prevent damage to the guidewire but has poor puncturability and flexibility. The flexibility of a needle is another factor contributing to the success of the procedure. Cobalt-chromium is generally harder than stainless steel, so a needle made of stainless steel is easy to operate in a long scope position. The EZ Shot 3 Plus 19G (Olympus, Tokyo, Japan) is a Menghini-type needle made of a nitinol pipe and a coil-shape sheath, and it exhibits good flexibility during the EUS-TD in a high bending position.
The guidewire must have adequate stiffness to deliver accessories and the guidewire tip needs to have a proper angle not to penetrate the pseudo-wall of PFCs. There are two main diameters of guidewires: 0.035 or 0.025 inches. The 0.035-inch guidewire has good stiffness, but there is a risk of damage to the guidewire by the puncture needle tip bevel, as well as a risk of injury to the pseudo-wall during the manipulation of the guidewire. Compared to the 0.035-inch guidewire, the 0.025-inch guidewire has a lower risk of needle tip bevel damage and damage to the pseudo wall of the PFC. This guidewire also shows good rotation in the pseudocyst wall of PFCs, but it has the disadvantage of low stiffness. Recently, 0.025-inch or 0.035-inch guidewires with appropriate stiffness (VisiGlide 2; Olympus and Jagwire; Boston Scientific, Natick, MA, USA) have been developed, increasing the convenience of the procedure.
Dilatation of the fistula tract is required to insert a stent into a PFC. In general, there are two methods of dilatation: cold dilatation (bougie or dilating balloon catheter) or electrocautery dilatation (cystotome or needle knife). In EUS-TD for PFCs, the bleeding risk is not higher than the procedure for other organs. Therefore, when inserting a metal stent, electrocautery dilatation can shorten the procedure, reduce the fluid leakage through the fistula tract, and lower the risk of fluid aspiration into the lungs. When performing electrocautery dilatation, a cystotome is slightly more advantageous than a needle knife because the direction of a cystotome coincides with the direction of the guidewire, unlike a needle knife. If a plastic stent is to be inserted, the fistula tract needs to be dilated sufficiently. Therefore, additional dilatation should be attempted using a 6-mm or 8-mm balloon dilation catheter or an 8-mm to12-mm CRE balloon after electrocautery dilatation. Recently, an ultra-slim 3-Fr balloon dilatation catheter (REN; Kaneka Medix, Osaka, Japan), which has tapered tips, has been developed, enabling tract dilatation without electrocautery.23
In general, the working channel diameter of a linear echoendoscope is about 3.7 to 3.8 mm, so a 7- to 8-Fr plastic stent is typically used. A double pigtail plastic stent without a side hole is used to prevent inner or outer migration of the stent and to prevent leakage of cystic fluid into the peritoneum (Fig. 5). The length of the stent should be selected to provide an additional length of at least 2 cm between the farthest end of the PFC and the intestinal lumen.24 Plastic stents are easy to remove, can be deployed for a long period, and have the advantage of being cheaper than metal stents. However, due to the small diameter of the stent, stent occlusion, and ineffective drainage, the need to perform re-intervention can occur. Therefore, multiple stenting may be required for proper drainage. However, inserting multiple stents has disadvantages, such as having sufficient fistula tract dilation, the technical difficulty of the procedure, and the prolonged procedure time compared to metal stents.
Compared to a plastic stent, an FCSEMS has a larger diameter, which can make the drainage more effective and can reduce the likelihood of stent obstruction. Moreover, the procedure is simpler and requires less time due to omitting the additional fistula tract dilation. However, an FCSEMS is more expensive than a plastic stent, so it is important to determine the indication well. Available metal stents include tubular metal stents, modified tubular metal stents, and lumen-apposing metal stents (LAMSs) (Fig. 6). Initially tubular metal stents were used frequently. A tubular metal stent has no flange to fix to the wall, and it is not very different from the metal stent that is used for a bile duct drain. A tubular metal stent has a low risk of tissue trauma, delayed bleeding, or perforation, but has a high risk of migration. There are a few studies on the use of tubular stents for PFCs, but these studies have not consistently reported that tubular stents are superior to plastic stents. In a retrospective study comparing biliary FCSEMSs and plastic stents in 230 patients, the FCSEMS group had a higher PFC resolution rate, while the plastic stent group had 2.9 times as many adverse events.25 In other retrospective studies comparing FCSEMSs and plastic stents in WOPN patients, the FCSEMS groups had significantly higher success rates.26,27 However, in a study by Mukai et al,28 there were no significant differences in technical success, clinical success, and adverse events between the FCSEMS group and the plastic stent group, and the mean procedure time was shorter in the FSCEMS group. If the PFCs are large, accompanied by infection or necrotic tissue inside, a metal stent is expected to be more helpful than a plastic stent, but a well-designed prospective study is needed to confirm this possibility. A modified tubular metal stent (BONA-Soo stent; Standard Sci Tech Inc., Seoul, Korea) has been developed with a modified distal end to overcome the migration disadvantage of the tubular metal stents, and its usefulness is being reported.29
A LAMS is an FCSEMS designed for EUS-guided drainage. LAMSs are produced by several different companies, including the NAGITM stent (TaeWoong Medical Co., Ltd., Ilsan, Korea), the AXIOSTM stent (Boston Scientific), the Niti-S SPAXUSTM stent (TaeWoong Medical Co., Ltd.), the Aixstent®PPS (LeufenMedical, Aachen, Germany), and the Hanarostent® PlumberTM (M.I. Tech., Seoul, Korea).30 A LAMS has flanges at both ends, a diameter of 10–20 mm, and a short length of 10–30 mm. The method of inserting these stents is similar to that of other FCSEMSs, and their radiopaque marker can help during deployment. Compared to a plastic stent, a LAMS has a short procedure time and larger drainage lumen that offers the advantage of a low risk of stent occlusion. In particular, a LAMS with a diameter ≥ 15 mm can pass through the gastroscope, so necrosectomy can be performed. A LAMS is covered with a silicone membrane that minimizes the ingrowth of the tissue into the stent and allows the stent to be easily removed. Due to these advantages, a LAMS is preferred for WOPN containing a large amount of necrotic material (Fig. 7). Recently, Niti-S hot SPAXUS stents and hot AXIOS stents equipped with electrocautery devices have been developed, making it possible to insert a stent along a guidewire after needle puncture without tract dilation. In addition, if the scope position of the linear echoendoscope is stable in the PFCs, the stent can be inserted by directly puncturing the cystic fluid with an electrocautery device embedded in the stent without a needle puncture. Itoi et al31 conducted a pilot study in which LAMSs were used in 15 symptomatic PP patients. In this study, 100% clinical and technical success was reported, and there was no recurrence for 11 months. One patient showed stent migration, but there were no clinical sequelae. Subsequent studies using LAMSs have reported high technical success rates and good clinical efficacy.32–34 However, in a retrospective case-control study by Bang et al35 for WOPN and PP patients, there was no significant difference in the technical success rate between LAMSs and plastic stents, and the procedure time was shorter with the LAMSs. Despite the advantages of LAMSs (shorter procedure time, a potential reduction in the occurrence of adverse events, and a lower risk of radiation exposure), the authors also raised questions about whether it is appropriate to use LAMSs, which are three times more expensive, for PFC drainage in all cases.
For successful EUS-TD with a LAMS, the size and shape of the PFC should be identified before the procedure. If the PFC is small, there may not be enough space for the inner flanges of the LAMS to unfold. For example, to insert a 10-mm-diameter SPAXUS stent, because each flange length before deployment is 25 mm, the depth of the cystic fluid must be at least 5–6 cm so that the stent can be easily deployed without impacting the opposite wall.36 In addition, even if the overall size of PFC is large, but it has a vertical length of less than 2 cm, it may be more appropriate to insert multiple plastic stents than a LAMS.26
Direct endoscopic necrosectomy (DEN) is a step-up approach that can be selected when antibiotics and transmural drainage are not effective. To start the procedure, if a double pigtail stent has been inserted as an initial treatment, the opening area is widened up to 15 mm with a large balloon dilation. Then, an FCSEMS with a diameter of 15 mm or more is inserted to enable the gastroscope to be inserted into the WOPN through the opening area. Next, small debris is removed with irrigation and suction, and solid materials are carefully removed from the WOPN cavity using tools such as retrieval baskets (Gemini, Microvasive; Boston Scientific, Marlborough, MA, USA) or retrieval nets (Fig. 8). If a LAMS with a large diameter is used, small debris in the WOPN cavity can flow out naturally, thus reducing the number of DEN procedures and making it easier to insert a gastroscope into the WOPN cavity at the next DEN procedure. The LAMS is removed before tissue overgrowth or membrane breakdown, as soon as the PFC with necrotic materials resolves.37 It is recommended to use CO2 instead of air to avoid gas embolism during DEN. In addition, hydrogen peroxide irrigation has been used to promote debridement.
Seewald et al38 first introduced a method of removing debris by inserting a gastroscope into the necrotic cavity through the fistula tract. In this study, 91% of 13 patients showed WOPN resolution, while two patients showed recurrence of the disease and underwent surgery. A large multicenter study of 93 patients conducted by Seifert et al39 reported an 80% clinical success rate, a 23% complication rate, and a 7.5% mortality rate. In the 2011 study of Gardner et al40 with 104 WOPN patients, WOPN resolution was observed in 91% of patients, with a complication rate of 14%. There has been some concern that the DEN procedure itself may increase the risk of super-infection, and clear timing and indications for DEN have not yet been established. Therefore, it is necessary to conduct additional studies on DEN.
The multiple transluminal gateway technique (MTGT) is step-up approach that creates several access routes under EUS in the stomach or duodenum for more effective drainage of WOPN. Varadarajulu et al41 were the first to describe the MTGT and reported a higher treatment success rate in patients with MTGT than in those with conventional drainage (91.7% vs. 52.1%,
The number of complications related to EUS-TD for PFCs is decreasing significantly with advances in procedures and the development of dedicated devices and metal stents. Procedure-related complications can occur in 5% to 30% of patients, and the most significant complications are bleeding and perforation.44–47
Bleeding has been reported in 1% to 10% of patients who underwent EUS-TD.44,48 Bleeding can occur at the needle puncture site, during the tract dilatation or debris removal. It can also occur due to mucosal injury by the metal stent, a pseudoaneurysm formation, or stent removal. To reduce the risk of bleeding, the patient's coagulation parameters should be measured before the procedure. Whether the patient is taking anticoagulants, antiplatelets, or others should be checked in advance. EUS-TD must be performed while checking the intervening vessel using a color Doppler before puncture. For a patient with a LAMS, it is necessary to check the resolution of the PFC 3 to 4 weeks after the procedure and, if the resolution is verified, it is necessary to consider removing the LAMS within 4 weeks.49,50 If bleeding occurs, attempts should be made to stop the bleeding using such methods as coagulation, epinephrine injection, balloon tamponade, and FCSEMS insertion. If the bleeding does not stop even with these methods and massive bleeding continues, trans-arterial embolization or surgical treatment should be considered. In a systematic meta-analysis of 455 patients in 14 studies related to DEN, bleeding was found to be the most common complication.47 Among the patients with bleeding, 93% responded to endoscopic hemostasis, but 7% needed angiography and embolization of the pseudoaneurysm (Fig. 9).
Perforation occurs in 0% to 4% of patients, causing very serious consequences.44,51 Perforation usually occurs by separation of the intestinal wall and cystic wall during the procedure or by cyst wall perforation during the exchange of the guidewire or stent. It can also occur due to the use of a non-coaxial needle knife or during the dilatation to a large balloon diameter. If a perforation occurs due to a defect between the intestinal wall and the cystic wall, it is necessary to consider inserting a LAMS. If the deployed stent is placed in the retroperitoneum or if peritonitis symptoms occur due to cystic fluid content, immediate surgery should be considered. However, if there are no peritonitis symptoms and the patient's vital signs are stable, conservative therapy such as fasting, antibiotics, and fluid therapy can also be considered. In addition, buried LAMS, development in chronic pancreatitis, and air embolism may occur.
EUS-TD has been established as an important treatment method for PFCs, which significantly influence the prognosis of acute pancreatitis. However, other treatment options including percutaneous or surgical approach or combination therapy may be considered depending on the characteristic of PFCs and the patient’s condition. More dedicated devices for EUS-TD should be developed, and additional well-designed multicenter prospective studies are required to resolve ongoing debates over EUS-TD.
None.
No potential conflict of interest relevant to this article was reported.
© The Society of Gastrointestinal Intervention. Powered by INFOrang Co., Ltd.