Endoscopic Management of Pancreaticobiliary Emergencies


 Gallstone pancreatitis

 Infected pancreatic pseudocyst

 Infected pancreatic necrosis

 Occluded pancreatic stent


 Acute cholangitis

 Acute cholecystitis

 Bile leak

 Hepatic abscess


In this section, we address acute gallstone pancreatitis, infected pancreatic pseudocysts, infected pancreatic necrosis, and occluded pancreatic stents. Of note, these entities are generally considered endoscopic “emergencies” to the extent that there is at least clinical suspicion or evidence of sepsis or similar decompensation, which can be alleviated endoscopically.

Acute Gallstone Pancreatitis

Gallstones represent the most common cause of acute pancreatitis , accounting for approximately 35–45 % of cases worldwide [3]. The mechanism by which gallstones induce acute pancreatitis is uncertain, but it may be related to reflux of bile or to pancreatic ductal hypertension from obstruction at the ampulla (due to the presence of a stone or to ampullary edema consequent to recent passage of a stone) [4]. Although a majority of patients will have a brief and conservatively manageable disease course, up to 25 % will develop local and/or systemic complications requiring more intensive management; this risk is higher among those with necrotizing pancreatitis as compared to acute interstitial edematous pancreatitis.

The role and timing of endoscopic intervention in the management of acute gallstone pancreatitis has been a subject of debate for several decades. Randomized clinical trials, systematic reviews, and meta-analyses have, over the years, provided conflicting results [59]; this is likely related to heterogeneity in patient samples, study methodology, definitions (e.g., of what constitutes “early”), and endpoints. As a result, large prospective studies continue to be performed to settle controversies and perhaps more carefully identify subgroups which may particularly benefit (or not) from endoscopic intervention.

In a recent Cochrane review and meta-analysis, it was determined that early endoscopic retrograde cholangiopancreatography (ERCP) did not significantly affect mortality and local or systemic complications of pancreatitis, regardless of predicted severity [10]. Only in the subgroup of patients with acute cholangitis or biliary obstruction were improved outcomes seen with ERCP. These findings are congruent with the most recent societal guidelines, which provide relatively clear recommendations in this regard [11]. Per consensus, urgent ERCP is not needed in most patients with gallstone pancreatitis in the absence of laboratory or clinical evidence of ongoing biliary sepsis or obstruction [5]. In cases where the index of suspicion is high for choledocholithiasis despite the absence of findings suggestive of biliary obstruction or sepsis, magnetic resonance cholangiopancreatography (MRCP) or endoscopic ultrasound (EUS) should be pursued in lieu of ERCP. For the minority of patients in whom ERCP is indicated (i.e., those with acute pancreatitis and concurrent acute cholangitis), ERCP is recommended within 24 h of presentation. Moreover, among this subset of patients, measures should be taken to prevent further exacerbation of pancreatitis post-ERCP (e.g., guidewire cannulation, pancreatic duct stents, and/or post-procedure rectal nonsteroidal anti-inflammatory drug suppositories) in high-risk patients even though they are not well studied in the context of acute gallstone pancreatitis [1113]. Until more data become available, the role of endoscopic management in acute gallstone pancreatitis thus appears to be limited to those with suspected concomitant acute cholangitis or biliary obstruction.

Infected Pancreatic Pseudocysts

Pancreatic pseudocysts develop in 10–20 % of patients following acute interstitial pancreatitis. Unlike acute fluid collections, pseudocysts are more mature collections (typically developing at least 4 weeks after acute pancreatitis) with generally homogeneous liquid components contained within a well-defined wall [14]. The management of pseudocysts depends primarily on the presence of symptoms and complications (e.g., infection) and to a lesser extent on the characteristics and location of the fluid collections. Asymptomatic pseudocysts, regardless of size or location, do not typically warrant intervention. For symptomatic (e.g., abdominal pain, fever, early satiety, jaundice) pseudocysts, drainage is critical, and while historically it was only a surgical option, the latter has been largely replaced by endoscopic and/or percutaneous (interventional radiologic) drainage. Each approach has its own advantages and disadvantages, and there is a paucity of comparative trials to determine the superiority of one technique over another. Therefore, the choice of drainage procedure is largely determined based on local expertise and the anatomical features of the fluid collection. In our experience, however, a majority of pseudocysts that abut the stomach or duodenum can be successfully drained endoscopically, and this has been demonstrated in numerous published reports.

Drainage, together with antibiotic therapy, becomes an urgent procedure in patients with signs of clinical deterioration due to suspected infection. The endoscopic approaches for drainage of pseudocysts are transpapillary, transmural, or a combination thereof [15]. The choice of approach is based upon the anatomic relationship of the pseudocyst to the stomach or duodenum, the presence of pancreatic ductal communication with the pseudocyst, and the size of the pseudocyst [15]. With respect to transmural drainage, this is achieved by large-bore stenting (plastic or metal) through the gastric or the duodenal wall [1517]. Pre-drainage imaging with EUS has been advocated to limit complications, but lack of EUS availability need not preclude transmural drainage if pre-procedural computed tomography (CT) imaging is available. The main advantages of EUS-guided drainage are the ability to detect unsuspected perigastric varices or other vascular structures and to facilitate transmural drainage in the absence of an endoscopically visible bulge, particularly for lesions near the pancreatic tail [18]. EUS-guided drainage should thus be employed in select circumstances, such as a small “window” of entry based on pre-procedural CT findings, unusual pseudocyst location, or documented major intervening vasculature.

Infected Pancreatic Necrosis

Approximately 10 % of patients with acute pancreatitis develop necrosis of the pancreatic parenchyma, the peripancreatic tissue, or both [14]. The most common manifestation of necrotizing pancreatitis is necrosis involving both the pancreas and peripancreatic tissues. The natural history of necrotizing pancreatitis is variable as the necrotic collection(s) may remain solid or liquefy, be localized or extend into the pelvis and elsewhere in the abdomen, and/or become infected, persist, or spontaneously disappear over time.

Infected pancreatic necrosis, in itself, is not an indication for emergent endoscopic intervention. Indeed, in stable patients with infected necrosis, initial treatment is expectant (including antibiotics), as surgical, radiologic, and/or endoscopic drainage should be delayed to allow for (semi) liquefaction of the contents and development of a fibrous wall around the necrotic material (i.e., walled-off pancreatic necrosis [WOPN]); this process generally takes at least 4 weeks, and in some cases, as with pseudocysts, necrotic collections can spontaneously resolve. By this point in time, if a drainage procedure is still needed, it is generally planned via a multimodality (endoscopic, radiologic and/or surgical) approach and on an elective basis. However, drainage may be considered on a more urgent basis in the uncommon circumstance wherein conservative management for 4 weeks is not deemed feasible because of ongoing symptoms and/or clinical decompensation related to infection (Fig. 18.1). Intervention in these settings carries greater risk (e.g., perforation, leakage) in part due to the lack of maturity of the acute necrotic collection as well as the generally higher level of patient morbidity. Whenever possible, an aspirate of the collection should be obtained and sent for microbiological studies to guide antibiotic therapy.


Fig. 18.1
Infected pancreatic necrosis with symptomatic mass effect including biliary and gastric outlet obstruction on CT. Endoscopic necrosectomy at 3 weeks post-pancreatitis resulted in clearance of infection and improvement of mass effect

As with pseudocysts, endoscopic drainage for immature necrotic collections or for WOPN can be performed through the stomach or duodenum, depending on the location of the necrotic collection. Because necrotic tissue is of variable texture and frequently includes a gelatinous, heterogeneous, and semisolid component, standard drainage with stenting (as employed for pseudocysts) is often inadequate, and direct endoscopic necrosectomy is generally needed [19]. Even necrosectomy, however, may require repeated endoscopy for lavage, additional tract dilation, stenting, debridement, and other interventions. At the end of endoscopic treatment, outcomes of necrosectomy in centers with expertise are favorable, reaching success rates of approximately 90 % [20]; nevertheless, it is important to discuss with the patient and care team, prior to the first endoscopic procedure, the potential requirement for multiple endoscopy sessions to achieve therapeutic success.

Occluded Pancreatic Stents

Pancreatic duct stent placement has been increasingly used for the treatment of a variety of disorders including chronic pancreatitis, pancreatic duct leaks or disruptions, drainage of pseudocysts, and the prevention of post-ERCP pancreatitis. Pancreatic duct stent occlusion is a potential late complication, which occurs at rates similar to those described for biliary stents of the same caliber. Most (small caliber) pancreatic stents occlude within 2 months [21]. Stent occlusion, similar to the biliary tree, is related to adherence of protein matrices to the inner surface of the stent, with or without interspersed mixed bacterial flora and calcium carbonate precipitates (Fig. 18.2) [22, 23]. Stent occlusion may be silent or may lead to increased pain, acute pancreatitis, or other acute and potentially lethal complications [24, 25]. In these circumstances, endoscopic removal or replacement of the pancreatic stent may be urgent or even emergent.


Fig. 18.2
Pancreatic duct obstruction . The patient had undergone balloon dilation of a pancreatic duct stricture with plastic stent placement; within 1 week, the patient developed acute pancreatitis. ERCP was performed, and the pancreatic duct stent was removed. (a) Pancreatogram revealed ductal irregularity, filling defects, and proximal side branch dilatation. (b) Balloon dilation performed. (c) Ductal sweeping revealed numerous calcium carbonate pancreatoliths

To avoid complications related to pancreatic duct occlusion, pancreatic stents should be removed or exchanged electively at predetermined intervals. Although the duration of this interval has not been determined (and may well depend on the initial indication for stent placement), an indwelling period of 2–6 weeks appears reasonable; covered metal stents may be left in situ for longer periods [26].


In this section, the topics of acute cholangitis, acute cholecystitis, bile leaks, and liver abscesses are addressed. As with the previously discussed pancreatic disorders, not all of these biliary disorders require emergent endoscopic intervention, and indeed some may have alternative management options, be it medical (e.g., antibiotics) or procedural (e.g., interventional radiology).

Acute Cholangitis

Acute cholangitis is a state of biliary infection associated with partial or complete obstruction of the biliary tree caused by any of various etiologies [27, 28], including choledocholithiasis (most common), benign and malignant strictures or masses, and indwelling biliary stent malfunction, among others. From a pathophysiological perspective, biliary infection alone is not believed to cause clinical cholangitis unless biliary obstruction raises the intraductal pressure sufficiently to cause cholangiovenous or cholangiolymphatic reflux [29]. The diagnosis of acute cholangitis is made on the basis of clinical findings such as Charcot’s triad, first described in 1877 [30], or Reynold’s pentad, first described in 1959 [31], in combination with serum biochemical data and imaging findings. A variety of different criteria for acute cholangitis have been historically used in the published literature, each with varying (but generally low) sensitivity and specificity; however, consensus definitions have been established [32] and recently revised [33].

Rapid and precise determination of the cause and severity of acute cholangitis is critical for appropriate management. Consensus criteria have been proposed for defining the severity of acute cholangitis and consist of three grades [28]. In brief, these consist of:


Severe, which is characterized by the onset of dysfunction or failure of at least one organ system despite supportive care with intravenous antibiotics and fluid resuscitation



Moderate, which entails at least two risk factors for progression to organ dysfunction (e.g., temperature ≥39 °C and serum bilirubin ≥5 mg/dL)



Mild, which does not meet criteria for either of the other two grades


The timing of biliary drainage is based on the severity grade [34]; urgent or emergent drainage is indicated in severe disease, while “early” drainage (generally defined as <72 h) is recommended for moderate disease [3436].

Endoscopic drainage with ERCP , whenever feasible, is generally advocated over surgical and percutaneous drainage [37]. The endoscopic approach to drainage, once selected as the modality of choice in a given case, can vary depending on the underlying etiology of acute cholangitis. Considerations for a few of the common etiologies of acute cholangitis are described below.


The standard approach to choledocholithiasis is endoscopic sphincterotomy followed by stone extraction with an occlusion balloon or basket. This allows for successful extraction of >95 % of stones <1.5 cm when performed by experienced endoscopists and in the absence of underlying strictures or altered bilioenteric anatomy [38]. Larger stones may require more advanced techniques, which are discussed later. The requisite length of sphincterotomy depends on papillary anatomy and stone size, and in some instances (e.g., when stone size is small and/or the papilla is patulous), it may not be necessary. There are also select clinical scenarios where sphincterotomy alone is insufficient to allow extraction of choledocholithiasis, in which case combined endoscopic papillary balloon dilation may be preferable [39, 40]. The use of prophylactic pancreatic duct stents and/or nonsteroidal anti-inflammatory drugs should be considered in these patients to decrease the risk of associated post-ERCP pancreatitis [1113].

When a stone is anticipated but not visualized during cholangiography, the potential advantages of empiric endoscopic sphincterotomy (e.g., for facilitating bile duct sweeping and increase detection of small stones) [41] should be weighed against the potential short- and long-term complications of an unnecessary sphincterotomy. Ultimately, the decision to perform or forego sphincterotomy is at the discretion of the endoscopist and influenced by the pretest probability for choledocholithiasis, quality of fluoroscopy, and availability of potentially helpful ancillary techniques (e.g., EUS). Of note, when incomplete stone extraction is known or suspected, a biliary stent should be placed to aid in biliary drainage [42].

Extraction of impacted and large bile duct stones may require additional techniques for successful removal and biliary drainage [43]. Stones impacting the ampulla make traditional biliary cannulation and sphincterotomy difficult or unfeasible; in these cases, needle-knife sphincterotomy is generally effective in disimpacting the stone, and the underlying stone may in fact help protect the pancreatic sphincter from inadvertent thermal injury from the needle knife (Fig. 18.3) [44, 45]. Thereafter, lithotripsy may be required for extraction of residual large stones. One form of lithotripsy is mechanical lithotripsy , in which a stone is captured in a specialized large basket and crushed (Fig. 18.4) [46]; fragments are then extracted using standard techniques. Another form of lithotripsy is intraductal lithotripsy , in which laser or electrohydraulic lithotripsy catheters are passed into the bile duct and used to fragment stones under direct endoscopic visualization via choledochoscopy [47]. Direct visualization is essential (unless a “centering” balloon is used [48]) to ensure that the lithotripsy catheter is directed at the stone and not the bile duct wall so as to prevent choledochal injury. An ancillary technique for large stone management is the combination of endoscopic biliary sphincterotomy and large-diameter (>12 mm and up to 20 mm) papillary balloon dilation; since the initial description of this technique in 2003 [49], several studies have found this approach to be safe, to facilitate large stone extraction, and to decrease the need for mechanical lithotripsy [35]. Although beyond the scope of this chapter, extracorporeal shock wave lithotripsy represents another adjunctive modality in the management of endoscopically challenging choledocholithiasis [35].


Fig. 18.3
Common bile duct stone-associated acute cholangitis . (a) A radiolucent, partially obstructing choledocholith was seen during ERCP. (b) Biliary sphincterotomy was performed, and balloon sweeping successfully removed a single, large choledocholith


Fig. 18.4
Distal common bile duct stone . (a) EUS image of a 6.8 mm echogenic, obstructive choledocholith. (b) The same stone is seen during ERCP. (c) Mechanical lithotripsy is performed with a flower basket. (d) Cholangiogram following lithotripsy and balloon sweeping

Biliary Strictures

Biliary strictures can occur due to a variety of benign and malignant causes, but in nearly all instances, endoscopic management in the context of acute cholangitis consists of expeditious biliary drainage with balloon dilation (Fig. 18.5) and, if indicated, stenting of the stricture. As with management of biliary strictures in the absence of acute cholangitis, the type and number of stents placed (as well as the need for sphincterotomy) depend on a variety of factors, including the etiology of the stricture, location within the biliary tree (e.g., distal versus hilar), nature of prior endoscopic or non-endoscopic biliary therapies, and local expertise, among others [43]. In the presence of acute cholangitis, however, a distinction (as compared to elective biliary drainage of hilar obstruction) that should be made is that complete bilateral drainage should be performed; this is particularly the case if both sides of the biliary tree have been previously instrumented (and thus potentially contaminated) or if multifocal strictures (e.g., posttransplant non-anastomotic strictures [50], primary sclerosing cholangitis) are present. It is worth noting that while drainage is critical, the need for stent deployment (as compared to biliary balloon dilation and sweeping only) is less clear and may depend on the etiology of the underlying stricture and response to balloon dilation [5154].


Fig. 18.5
Acute cholangitis secondary to hepaticojejunostomy site biliary obstruction. (a) Endoscopic image of strictured hepaticojejunal anastomosis. (b) ERCP with balloon dilation of the hepaticojejunal stricture. (c) Cholangiogram following successful balloon dilation. (d) Endoscopic view of the hepaticojejunostomy post-dilation

Only gold members can continue reading. Log In or Register to continue

May 30, 2017 | Posted by in GASTROENTEROLOGY | Comments Off on Endoscopic Management of Pancreaticobiliary Emergencies
Premium Wordpress Themes by UFO Themes