Endoscopic management of postcholecystectomy biliary complications

INTRODUCTION — Symptomatic gallbladder disease is common and is typically managed with cholecystectomy. Laparoscopic cholecystectomy has been associated with higher rates of biliary complications than open cholecystectomy. This may be related to technical constraints of the minimally invasive approach. Endoscopy-guided interventions are used to manage some postcholecystectomy biliary conditions (eg, bile leaks from the cystic duct remnant or gallbladder bed). However, cholecystectomy-related injuries that transect the common bile or common hepatic duct generally require surgical intervention. (See "Repair of common bile duct injuries".)

Management of postcholecystectomy biliary injuries often involves a multidisciplinary team with input from biliary endoscopists, interventional radiologists, and hepatobiliary surgeons.

The focus of this topic is endoscopic management of biliary complications related to laparoscopic cholecystectomy.

The clinical features, diagnosis, and non-endoscopic management of cholecystectomy-related biliary injuries are presented separately. (See "Complications of laparoscopic cholecystectomy" and "Surgical common bile duct exploration".)

The management of acute cholecystitis and surgical techniques for laparoscopic cholecystectomy are discussed separately. (See "Treatment of acute calculous cholecystitis" and "Laparoscopic cholecystectomy".)

An overview of endoscopic retrograde cholangiopancreatography (ERCP) including preprocedure preparation is also discussed separately. (See "Overview of endoscopic retrograde cholangiopancreatography (ERCP) in adults".)

BILE LEAK

Incidence and etiology — Reported rates of bile leak following cholecystectomy range from 0.3 to 0.9 percent [1]. Bile leaks that are managed with endoscopic intervention usually originate from one of the following sites: the cystic duct remnant, the gallbladder bed, an accessory duct, or lateral injury to the common bile or hepatic duct [2,3].  

However, circumferential injury (ie, transection) to the common bile or common hepatic duct typically requires non-endoscopic assessment and repair, and this is discussed separately. (See "Repair of common bile duct injuries".)

Postcholecystectomy biliary injuries have been described in several classification systems, and we typically use the Hanover classification (table 1) [4-6]. Another commonly used system, the Strasberg classification, is outlined in the figure (figure 1). (See "Complications of laparoscopic cholecystectomy", section on 'Strasberg-Bismuth classification'.)

Preprocedure evaluation — Postcholecystectomy bile leaks are often suspected based on presenting symptoms (eg, abdominal pain, fever, jaundice) and laboratory studies (ie, cholestatic pattern of liver test abnormalities). The diagnosis of bile leak is confirmed with imaging that demonstrates a fluid collection (eg, transabdominal ultrasound, magnetic resonance cholangiopancreatography [MRCP]). In addition, hepatobiliary cholescintigraphy (generally referred to as a HIDA scan) may also be performed when other studies are equivocal. Thus, most patients will have had laboratory studies and imaging, thereby mitigating the need for additional testing before endoscopy-guided intervention. Review of the radiographic imaging prior to endoscopy will help to define the biliary ductal anatomy. (See "Complications of laparoscopic cholecystectomy", section on 'Bile leaks (type A, C, D injury)'.)

The clinical manifestations and diagnosis of biliary injuries related to laparoscopic cholecystectomy are discussed in more detail separately. (See "Complications of laparoscopic cholecystectomy", section on 'Biliary injury'.)

Endoscopic intervention

Goals — The goal of endoscopic retrograde cholangiography (ERC)-guided therapy for bile leak is to eliminate the transpapillary pressure gradient, thereby facilitating transpapillary flow of bile into the duodenum rather than extravasation at the site of the leak.

Timing — For most patients with a postcholecystectomy bile leak who require endoscopic intervention, we perform ERC within 12 to 24 hours after the bile leak is diagnosed because delaying the procedure may lead to a longer hospital stay, increased costs, and possibly increased risk for adverse events. (See "Complications of laparoscopic cholecystectomy", section on 'Bile leaks (type A, C, D injury)'.)  

While consensus guidance and randomized trials on timing for endoscopic intervention for bile leaks are lacking, retrospective studies have suggested that procedure timing was not associated with worse outcomes [7,8]. In a database study including over 1000 patients who underwent ERC for bile duct leak after cholecystectomy, timing of ERC (ie, within 24 hours, after two to three days, or after >3 days) was not associated with significantly different rates of post-ERC adverse events (eg, infection) [7]. However, the possibility of selection bias cannot be excluded, and more studies are needed to evaluate other outcomes, including length of hospital stay and cost.

Procedure technique

Localizing the injury site — Localizing most bile leaks and other abnormalities such as retained common bile duct stones can be achieved with ERC [9].

Bile leaks that are typically managed endoscopically originate from one of the following injuries (table 1) [5,10]:

●Peripheral biliary leakage from the cystic duct remnant (eg, related to impaired clip closure or pressure from retained gallstone (image 1)), the gallbladder bed, or an accessory duct (eg, duct of Luschka) (image 2A-B).

●Noncircumferential injury to the lateral aspect of the common bile, common hepatic, or right hepatic ducts.

●Transection without ligation of an aberrant hepatic duct, provided that a guidewire can bridge the leak and establish drainage of the right intrahepatic ducts (image 3).

Establishing biliary drainage — Selecting an intervention for treating the bile leak is informed by the presence of retained common bile duct stones:

●Bile leak without retained stones – For most patients with bile leak but without retained stones, we typically perform ERC-guided placement of a transpapillary plastic biliary stent (size range, 7 to 10 French) without biliary sphincterotomy because this approach is effective for treating bile leaks while avoiding the possible risks associated with sphincterotomy (eg, bleeding) (image 4) [11]. (See "Endoscopic biliary sphincterotomy", section on 'Complications'.)

Most patients experience improvement in symptoms and liver enzymes within 24 to 48 hours following stent placement. For patients with a percutaneous biliary catheter in place, external biliary drainage declines within the first 24 hours after the procedure.

●Bile leak with retained stone(s) – For most patients with retained common bile duct stone(s) and bile leak, we typically perform biliary sphincterotomy and stone extraction followed by biliary stent placement. Biliary sphincterotomy is needed prior to stone extraction with balloon or basket catheters. Removal of bile duct stones is discussed separately. (See "Endoscopic management of bile duct stones".)

For patients with a lower grade bile leak in the setting of retained stones, some endoscopists perform sphincterotomy and stone removal without stent placement to manage the bile leak [12]. Lower grade leaks have been described as leaks that require complete or near complete intrahepatic filling with contrast to demonstrate contrast extravasation (image 5) [10,12]. However, we typically place a stent to manage such bile leaks because stent placement is associated with high rates of successful leak closure. (See 'Outcomes' below.)  

Stent removal — The biliary stent remains in place for approximately four to six weeks. After this indwelling time, most patients have recovered clinically (ie, no abdominal pain/distension, fever, or jaundice) and the bile leak has healed. If a percutaneous drain had been placed, the drainage should be less than 10 mL per day prior to removing the stent.

For patients with a bile leak involving noncircumferential injury to the common bile or common hepatic duct, some clinicians obtain a HIDA scan in addition to assessing symptoms to confirm that the leak has resolved prior to stent removal. (See "Complications of laparoscopic cholecystectomy", section on 'Bile leaks (type A, C, D injury)' and 'Localizing the injury site' above.)

For patients with symptomatic improvement, we typically remove the stent by performing side-viewing endoscopy with cholangiography to confirm that the bile leak has resolved and to exclude the presence of sludge or retained bile duct stones [13,14]. (See "Overview of upper gastrointestinal endoscopy (esophagogastroduodenoscopy)".)

The approach to managing patients with persistent symptoms and nonresolving bile leak despite initial endoscopic intervention is discussed below. (See 'Nonresolving bile leaks' below.)

Adverse events — Adverse events associated with ERC-guided biliary stent placement may be related to the ERC or to stent placement:

●ERC-related – Adverse events associated with ERC may be due to the endoscopy itself (eg, pancreatitis) or due to anesthesia. These complications are discussed in more detail separately. (See "Overview of endoscopic retrograde cholangiopancreatography (ERCP) in adults", section on 'Adverse events' and "Anesthesia for gastrointestinal endoscopy in adults", section on 'Complications'.)

●Stent-related – Adverse events related to short-term biliary stent placement are uncommon but may include stent migration and cholangitis [8].

Data suggest that the risk of mortality is low for patients with bile leak who are treated with endoscopic intervention. In a study including 1028 patients with postcholecystectomy bile leak managed endoscopically, the mortality rate was 2 percent [7].

Outcomes — Short-term endoscopy-guided biliary stent placement is effective for treating bile leaks that do not involve circumferential injury of the common bile duct or common hepatic duct [7-9,11,15,16]. In an observational study including 1028 patients with postcholecystectomy bile leak, ERC-guided stent placement with or without sphincterotomy was associated with lower rates of persistent bile leak (defined as requiring surgical or percutaneous intervention) compared with sphincterotomy alone (3 and 4 percent, respectively, versus 11 percent) [7]. In a prior study including 100 patients with postcholecystectomy bile leak who had ERC-guided therapy, biliary stent placement was associated with lower rates of persistent leak requiring additional endoscopic intervention or surgery compared with sphincterotomy alone (3 versus 33 percent) [11]. In a small trial of 27 patients with a postcholecystectomy bile leak, biliary stent placement resulted in shorter time to leak closure compared with sphincterotomy alone (5 versus 10 days) among patients with nondilated common bile duct (≤8 mm in diameter) [15].

Nonresolving bile leaks — Some postcholecystectomy bile leaks do not resolve despite initial endoscopic intervention with plastic stent placement with or without sphincterotomy. Reported rates of nonresolving bile leaks that were confirmed with cholangiography have ranged from 3 to 8 percent [7,14,17].

●Clinical features and diagnosis – A nonresolving leak is suspected in patients with persistent symptoms (abdominal pain, elevated liver enzymes) and/or persistent percutaneous drainage after the initial endoscopic intervention.

Imaging with magnetic resonance cholangiopancreatography (MRCP), abdominal computed tomography scan, or percutaneous cholangiography (ie, for patients with a percutaneous drain) confirms the diagnosis of refractory bile leak, excludes anatomic variations, and informs the approach to management [18]. As an example, some refractory bile leaks originate from transection of an aberrant right hepatic duct, and this injury often requires surgical intervention. The bile leak occurs because the cystic duct arose from an aberrant right hepatic duct, and leakage originates from the distal aspect of what appears to be the cystic duct and the transected right duct branch. This injury is usually identified on MRCP. (See "Complications of laparoscopic cholecystectomy", section on 'Imaging'.)

●Endoscopic management – Options for endoscopic intervention for nonresolving bile leaks include:

•Biliary sphincterotomy and plastic stent exchange – For patients with refractory bile leak who were initially managed with stent placement only, we perform biliary sphincterotomy in combination with stent exchange. Endoscopic biliary sphincterotomy is discussed in detail separately. (See "Endoscopic biliary sphincterotomy".)

Stent exchange may involve placing a larger caliber plastic stent (10 French) and/or multiple plastic stents [19].

•Fully covered self-expandable metal stent – Placement of a fully covered, self-expandable metal stent (FCSEMS) is an option for bile leaks arising from the cystic duct remnant, common bile duct, or common hepatic duct that do not resolve with plastic stent placement [19-23]. Bile leaks that arise proximally to the confluence of the right and left hepatic ducts are not managed with FCSEMS because of the risk of obstructing drainage from the other side of the liver [10].

After placement, the FCSEMS is typically left in place for one to three months before endoscopic removal followed by cholangiography to confirm leak resolution. (See 'Stent removal' above.)

Observational studies suggest that FCSEMS are effective for achieving resolution of postcholecystectomy bile leaks [19-23]. In a study including 40 patients with postcholecystectomy bile leaks that did not resolve with placement of a 10 French transpapillary biliary stent combined with biliary sphincterotomy, FCSEMS placement was associated with higher rates of leak closure compared with multiple plastic stents (100 versus 83 percent) [19]. Mild pancreatitis was more commonly reported in the FCSEMS group (10 versus 0 percent), but it resolved with medical management. In other studies, adverse events with FCSEMS included stent migration, stent occlusion, and bile duct stricture [20,21].

For bile leaks that do not resolve with endoscopic management, surgical intervention (eg, hepaticojejunostomy) is usually required for long-term resolution. (See "Complications of laparoscopic cholecystectomy", section on 'Management'.)

BILIARY STRICTURE

Incidence and etiology — The reported rate of biliary stricture related to cholecystectomy is approximately 0.5 percent [24].

Some patients with a cholecystectomy-related stricture or partially occlusive surgical clip are managed with endoscopic retrograde cholangiography (ERC)-guided stricture dilation and stent placement. However, patients with biliary obstruction from complete transection or clipping of the common bile or common hepatic duct usually require radiology-guided percutaneous drainage with stenting and/or surgical intervention (hepaticojejunostomy) (image 6). (See "Repair of common bile duct injuries", section on 'Transection of common bile duct'.)

Endoscopic therapy is usually not technically feasible for such an obstructive injury because a guidewire cannot be advanced through the lesion and into the biliary tree [25]. (See "Complications of laparoscopic cholecystectomy", section on 'Injuries to common bile or common hepatic ducts (type E injury)'.)

Preprocedure evaluation — The diagnosis of postcholecystectomy biliary stricture is often suspected based on presenting symptoms (jaundice, abdominal pain, pruritus), laboratory studies (cholestatic pattern of liver test abnormalities) and imaging (eg, transabdominal ultrasound and/or magnetic resonance cholangiopancreatography [MRCP] showing dilated intrahepatic ducts). Thus, most patients will have had laboratory studies and imaging, thereby mitigating the need for additional testing before ERC. Review of MRCP prior to endoscopy will help to define the biliary ductal anatomy. (See "Complications of laparoscopic cholecystectomy", section on 'Imaging'.)

Endoscopic intervention

Goals — The goals of endoscopic intervention are to increase the diameter of the biliary stricture as demonstrated on cholangiography, improve flow of bile through the stricture, and resolve symptoms and laboratory abnormalities associated with cholestasis (jaundice, elevated liver enzymes).

Procedure technique — During the initial endoscopic session, we perform the following interventions (image 7A-C) [26]:

●Place a guidewire across the stricture.

●Perform biliary sphincterotomy to facilitate stent placement and future stent exchanges (if needed).

●Dilate the stricture using balloons with diameters ranging from 4 to 10 mm. Selecting an initial balloon size is informed by the diameter of the proximal bile duct and endoscopist preference.

●Place a plastic biliary stent (size range: 7 to 11.5 French). For most patients, we place a 10 French plastic stent.

After the index procedure, ERC is typically repeated in three to six months for the following interventions: the previously-placed stent(s) are removed, the stricture is examined with cholangiography and dilated if it remains visible, and the new plastic stent(s) are replaced across the stricture if they are still warranted. Most patients require a total indwelling stent time of 6 to 12 months to achieve stricture resolution. The optimal number, size, and frequency of stent exchanges has not been established by consensus [27-29].

Adverse events — Adverse events reported with ERC-guided biliary stent placement may be related to the ERC, stricture dilation, or stent placement:

●ERC-related – Adverse events associated with ERC may be due to the endoscopy itself (eg, pancreatitis) or due to anesthesia. These adverse events are discussed in more detail separately. (See "Overview of endoscopic retrograde cholangiopancreatography (ERCP) in adults", section on 'Adverse events' and "Anesthesia for gastrointestinal endoscopy in adults", section on 'Complications'.)

●Stent-related – Adverse events associated with short-term biliary stent placement include stent migration, stent occlusion, cholangitis, and bowel perforation (rare) [30]. Stent dysfunction can typically be managed endoscopically with stent exchange.

Outcomes — For most patients with cholecystectomy-related biliary strictures, endoscopic intervention with biliary stent placement has been associated with long-term stricture resolution while avoiding external drain placement [27,30,31]. In a study of 96 patients with postcholecystectomy stricture of the common bile duct, serial ERC-guided stent placement was associated with stricture resolution in 79 patients (82 percent) after a mean follow-up of six years [27]. The mean number of plastic stents inserted during one endoscopic session was 1.9 ± 0.89 stents (range 1 to 4 stents). In a study of 154 patients with postcholecystectomy biliary stricture but without complete bile duct transection, serial ERC with placement of an increasing number of plastic stents over time was associated with stricture resolution in 149 patients (97 percent) during a mean treatment duration of 12 months [30]. The mean number of stents inserted during one session in a side-by-side configuration was 4.3 ± 1.6 stents. After a mean follow-up of 11 years, the rate of stricture recurrence was 9 percent.

Indirect data from studies including patients with biliary strictures related to liver transplantation suggested that initial endoscopic management for postsurgical biliary complications was effective and associated with lower mortality rates compared with surgery [32].

Nonresolving biliary stricture — For patients with a nonresolving common bile duct stricture despite serial ERC sessions with plastic stent exchanges for up to 12 months, endoscopic options include placement of a fully covered, self-expandable metal (FCSEMS). Use of FCSEMS for benign biliary strictures related to cholecystectomy is off label in the United States, although FCSEMS are approved for treating biliary strictures related to chronic pancreatitis. (See "Overview of the complications of chronic pancreatitis", section on 'Obstruction'.)

The decision to use FCSEMS for biliary stricture following cholecystectomy should be carefully made on an individual basis while balancing the potential benefits (eg, stricture resolution without surgical intervention) with potential adverse events (eg, stent migration) and costs.

For patients in whom FCSEMS placement is indicated, we typically use an 8 or 10 mm stent, depending on the initial diameter of the bile duct. The FCSEMS remains in place for three to six months, and then it is removed endoscopically.

Data on the use of FCSEMS for benign biliary strictures are emerging [21,33,34]. In a study including 18 patients with postcholecystectomy biliary stricture, FCSEMS placement with median indwelling time of 11 months (range 1 to 14 months) was associated with stricture resolution in 13 patients (72 percent) [33]. At five years after stent removal, 85 percent of patients with initial stricture resolution remained stent-free. Serious adverse events related to ERC with stent placement occurred in seven patients (39 percent) and included cholangitis (six patients) and pancreatitis (one patient), but all events resolved without long-term sequelae. Other possible adverse events include stent migration and stricturing at the stent margins [34].

For patients with postcholecystectomy biliary stricture that does not resolve with endoscopic stenting, surgical intervention may be required for stricture resolution. (See "Complications of laparoscopic cholecystectomy", section on 'Management'.)

Monitoring for recurrent stricture — After removing the stent(s) and confirming stricture resolution with ERC, we monitor patients periodically with liver enzymes (aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and total bilirubin; eg, every three months for 6 to 12 months). If symptoms recur or liver enzymes rise over baseline values, we obtain imaging (eg, MRCP or repeat ERC) to evaluate for stricture recurrence.

RETAINED COMMON BILE DUCT STONES — The approach to managing retained common bile duct stones following cholecystectomy (in the absence of bile leak) is similar to that for patients with an intact gallbladder and typically involves biliary sphincterotomy and stone removal with basket or balloon catheters. Removal of common bile duct stones is discussed in detail separately. (See "Endoscopic management of bile duct stones".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Biliary complications following surgery".)

SUMMARY AND RECOMMENDATIONS

●Background – Symptomatic gallbladder disease is common and is often managed with laparoscopic cholecystectomy. However, laparoscopic cholecystectomy has been associated with higher rates of biliary complications than open cholecystectomy (table 1). This may be related to technical constraints of the minimally invasive approach. (See 'Introduction' above and "Complications of laparoscopic cholecystectomy".)

●Bile leak – The goal of endoscopic therapy for postcholecystectomy bile leak is to eliminate the transpapillary pressure gradient, thereby facilitating transpapillary flow of bile into the duodenum rather than extravasation at the site of the leak.

For patients with a bile leak from a peripheral site (eg, cystic duct remnant) or from lateral injury to the common bile or common hepatic duct, we suggest endoscopic retrograde cholangiography (ERC)-guided placement of a transpapillary stent with or without biliary sphincterotomy rather than sphincterotomy alone because ERC-guided stent placement is more effective for resolving most bile leaks (Grade 2C). (See 'Bile leak' above.)

The biliary stent remains in place for approximately four to six weeks. After this stent indwelling time, most patients have recovered clinically (ie, no abdominal pain/distension), the bile leak has healed, and the stent is removed endoscopically.

●Biliary stricture – For most patients with postcholecystectomy biliary stricture in the absence of complete obstruction of the common bile or common hepatic duct, we suggest ERC-guided intervention rather than surgery (Grade 2C). An ERC-guided approach with biliary stent placement is less invasive and has been associated with long-term stricture resolution. (See 'Biliary stricture' above.)

ERC with routine plastic stent exchange is performed approximately every three to six months for a total stent indwelling time of 6 to 12 months. After stricture resolution, we monitor patients with liver enzymes every three months for 6 to 12 months and obtain imaging (eg, magnetic resonance cholangiopancreatography [MRCP]) when there is a rise in liver enzymes over baseline values.

●Retained common bile duct stones – The approach to managing retained common bile duct stones following cholecystectomy (in the absence of a bile leak) is similar to that for patients with an intact gallbladder and typically involves biliary sphincterotomy and stone removal with basket or balloon catheters. Removal of common bile duct stones is discussed in detail separately. (See "Endoscopic management of bile duct stones".)