sclerosing cholangitis, a rare disease of unknown cause, can result in multiple strictures of the

intrahepatic and extrahepatic bile ducts. This chapter focuses primarily on postoperative bile duct

strictures and primary sclerosing cholangitis.

POSTOPERATIVE BILE DUCT STRICTURES

Pathogenesis

1 Most benign bile duct strictures result from operations in or near the right upper quadrant. More than

80% of strictures occur after injury to the bile ducts during cholecystectomy. The exact incidence of bile

duct injury is unknown because many cases may go unreported in the literature. Data suggest that the

incidence of bile duct injury during open cholecystectomy is 1 in 500 to 1,000 cases. The incidence of

bile duct injury during laparoscopic cholecystectomy is clearly higher. Although a wide range in the

incidence of injury can be found in reported series, the most accurate data most likely come from

surveys encompassing thousands of patients. These reports reflect the results from a large number of

surgeons in both community and teaching hospitals. The results of such series suggest an incidence of

bile duct injury during laparoscopic cholecystectomy ranging from 0.3% to 0.7%.1 Furthermore, the

incidence of bile duct injury associated with laparoscopic cholecystectomy does not appear to have

diminished in more recent surveys, suggesting that the previously observed increase is not simply the

result of a learning curve associated with the laparoscopic technique. Finally, due to the high frequency

of laparoscopic cholecystectomy, it is estimated that one in every two or three surgeons will create a

bile duct injury during his or her career.

ETIOLOGY

Table 62-1 Causes of Benign Bile Duct Strictures

A number of factors are associated with bile duct injury during either open or laparoscopic

cholecystectomy, including acute or chronic inflammation, inadequate exposure, patient obesity, and

failure to identify structures before clamping, ligating, or dividing them. More specific causes of bile

duct injury also exist. Bleeding from the cystic or hepatic arteries can lead to bile duct injury during

attempts to gain hemostasis. The generous application of Ligaclips at either open or laparoscopic

cholecystectomy to hilar areas not well visualized can result in placing a clip on or across a bile duct,

with resultant injury (Fig. 62-1). Failure to recognize congenital anatomic anomalies of the bile ducts,

such as insertion of the right hepatic duct into the cystic duct or a long common wall between the cystic

duct and the common bile duct, can also lead to injury (Fig. 62-2). It should also be noted that a

significant discussion has ensued regarding the optimal timing of performing a laparoscopic

cholecystectomy in the setting of acute inflammation of the gallbladder. Although 30-day postoperative

morbidity and mortality rates may remain independent of timing, it is clear that patients who undergo

1605

laparoscopic cholecystectomy beyond 24 hours are more likely to require an open procedure, and

sustain significantly longer postoperative and overall lengths of hospital admission (and therefore

cost).2

A number of technical factors are associated with laparoscopic cholecystectomy that can also increase

the risk of bile duct injury compared with the open procedure. These factors include the use of an endviewing laparoscope, which alters the surgeon’s perspective of the operative field. The issue of visual

alignment and perspective has become even more topical with the proliferation of single incision

laparoscopic cholecystectomy which is known to be associated with a higher rate of common bile duct

injury than the traditional 4-incision laparoscopic technique utilizing an angled scope.3 Excessive

cephalad retraction of the gallbladder fundus can cause the cystic duct and common bile duct to become

aligned in the same plane. This distortion often results in the classic laparoscopic injury, in which the

common bile duct is mistaken for the cystic duct and clipped and divided (Fig. 62-3).4 The role of

intraoperative cholangiography (IOC) in preventing bile duct injury during laparoscopic

cholecystectomy is controversial. Individual series have failed to demonstrate that either performing

routine or selective IOC affects the incidence of bile duct injury. Although an initial retrospective

nationwide cohort analysis of Medicare patients undergoing laparoscopic cholecystectomy between

1992 and 1999 demonstrated that common bile duct injuries occurred in 0.39% of patients in which IOC

was performed versus 0.58% in patients not undergoing IOC (unadjusted relative risk, 1.49; 95%

confidence interval 1.42–1.57),5 a more recent Medicare-based study (2000 to 2009) analyzing over

92,000 patients undergoing cholecystectomy identified no statistically significant association between

IOC and common bile duct injury.6 The authors therefore concluded that IOC is not effective as a

preventive strategy against common duct injury during cholecystectomy.6 Furthermore, the proper

interpretation of IOC can minimize the extent of injury. Nevertheless, only 27% of surgeons in the

United States perform IOC routinely.7 Finally, ample evidence exists to support the conclusion that the

experience of the surgeon in performing laparoscopic cholecystectomy can be correlated with the risk of

bile duct injury.

Figure 62-1. Percutaneous transhepatic cholangiogram in a patient with a bile duct stricture secondary to iatrogenic injury during

cholecystectomy. Numerous surgical clips can be seen in the area of the stricture. (Reproduced with permission from Lillemoe KD,

Pitt HA, Cameron JL. Postoperative bile duct strictures. Surg Clin North Am 1990;70:1356–1380.)

1606

Figure 62-2. Operative cholangiogram demonstrating a right lobe segmental bile duct entering the cystic duct (arrow). Division of

the cystic duct proximal to this insertion can result in a bile leak or obstruction of bile flow from a significant segment of the liver.

Figure 62-3. Classic laparoscopic bile duct injury. The common bile duct is mistaken for the cystic duct and transected. A variable

extent of the extrahepatic biliary tree is resected with the gallbladder. The right hepatic artery, in background, is also often injured.

In recent years, there has been a growing understanding of surgeon cognitive factors associated with

bile duct injury during laparoscopic cholecystectomy. An analysis examining 252 biliary injuries during

laparoscopic cholecystectomy using human error factor and cognitive science techniques found that 97%

of injuries were caused by visual-perceptual illusion or inadequate visualization.8 Further work from the

same group has determined a major explanation for the surgeon’s frequent inability to recognize bile

duct injury. These bile duct injuries appear to be associated with confirmation bias, which is a

propensity to seek cues to confirm a belief and to discount cues that might discount the belief. Although

cognitive factors are important for the understanding of the psychological issues associated with bile

duct injuries, surgeons must continue to have the appropriate corrective mechanisms in place to

minimize the chance of these injuries, including knowledge of anatomy, typical mechanisms of injury,

and an appropriate level of suspicion and logic. An example of such a corrective mechanism occurs

within the operative technique of laparoscopic cholecystectomy, which defines the “critical view of

safety,” and therefore helps prevent misidentification and injury of the major bile ducts.9 This

anatomical-based safety concept can be further developed by including the orientation and relative

positions of the multiple structure of interest within adjacent regional anatomy (hepatic artery, sulcus of

Rouvier, umbilical fissure, and common bile duct).

The importance of ischemia of the bile duct in the formation of postoperative strictures has been

1607

emphasized. Injury to the hepatic artery at the time of biliary injury during laparoscopic

cholecystectomy has been recognized at an increased incidence, as high as 50%, when investigated at

the time of presentation.10 The true impact of an arterial injury however, remains debated. It is clear

that the most common site of vasculobiliary injury is the right hepatic artery.11 Damage to this vessel

can lead to a higher injury level on the bile duct than the gross observed mechanical injury.

Vasculobiliary injuries may also have specific effects on the arteries (pseudoaneurysm with delayed

hemorrhage), bile ducts (necrosis, stenosis, cholangitis) and/or liver (necrosis, atrophy) over variable

lengths of time.11 Finally, concurrent hepatic artery and portal vein injuries can have catastrophic

effects on the liver, including rapid necrosis. A more clinically common cause of ischemia can be

unnecessary dissection around the bile duct during cholecystectomy or bile duct anastomosis, which can

divide or injure the major arteries of the bile duct that run in the 3-o’clock and 9-o’clock positions.

Another important factor contributing to the formation of biliary strictures is the intense connective

tissue response with fibrosis and scarring that can occur after bile duct injury. Experimental studies of

bile duct ligation in a canine model have demonstrated immediate and sustained elevation of bile duct

pressure and progressive increase in bile duct diameter. Histologic changes at 1 month after ligation

have shown that the bile duct wall is thickened, with a reduction of mucosal folds and loss of surface

microvilli, associated with a well-defined epithelial degeneration. Biochemical analysis of connective

tissue response to ligation showed that collagen synthesis and prolene hydroxylase activity is increased

within 2 weeks in the obstructed bile duct and is sustained throughout the period of observation.

Finally, a marked local inflammatory response can develop in the adjacent tissue in association with bile

leakage, which occurs with many bile duct injuries. This inflammation can be further intensified in the

face of infection. This inflammation results in fibrosis and scarring in the periductal tissue, further

contributing to stricture formation. These factors can be of major importance in bile duct injuries during

laparoscopic cholecystectomy, which are frequently associated with bile leaks.

After cholecystectomy and common bile duct exploration, the two most common operations

associated with bile duct injury are gastrectomy and hepatic resection. The most common situation

resulting in bile duct injury during gastrectomy involves dissection of the pyloric region and the first

portion of the duodenum in the face of inflammation from peptic ulcer disease. The injury occurs during

mobilization of the duodenum either for creation of a Billroth I gastroduodenostomy or for closure of

the duodenal stump. Biliary injury during liver resection is most likely to occur during dissection of the

hepatic hilum.

In addition to iatrogenic bile duct injury occurring during cholecystectomy or other operations, bile

duct strictures can also occur at biliary anastomoses. Such strictures can occur at a biliary-enteric

anastomosis performed for reconstruction after resection for benign or malignant disease of the

pancreaticobiliary system, or after end-to-end bile duct anastomosis performed for hepatic

transplantation (even in the context of chronic immunosuppressive therapy) or for repair of traumatic

injury. Ischemia of the anastomosis caused by excessive skeletonization of the duct in preparation for

the anastomosis is an important factor in many such strictures.

1608

Algorithm 62-1. Algorithm for diagnosis and management of bile duct injury associated with laparoscopic cholecystectomy.

Unfortunately, the recurrence of bile duct strictures after an initial attempt at repair is not uncommon

and can also account for a number of anastomotic strictures.12,13 A number of other factors have been

evaluated in patients who have a recurrent bile duct stricture, including the location of the stricture, the

length of follow-up, the influence of previous operations, the type of operation performed, the type of

sutures used, and the use and duration of postoperative stenting.14 Previous attempts at repair,

performance of a procedure other than choledochojejunostomy or hepaticojejunostomy, and stricture

location higher in the biliary tree appear to be associated with a higher incidence of recurrent stricture.

Finally, long-term follow-up of a bile duct anastomosis is important because strictures can develop years

after the original anastomosis.

Clinical Presentation

2 Most patients with biliary injuries present early after their initial operation (Algorithm 62-1). After

open cholecystectomy, only approximately 10% of postoperative strictures are actually suspected within

the first week, but nearly 70% are diagnosed within the first 6 months, and more than 80% are

diagnosed within 1 year of surgery. In series reporting bile duct injuries during laparoscopic

cholecystectomy, the injury is usually recognized either during the procedure (25% to 30%) or, more

commonly, in the early postoperative period.

Patients suspected of having a postoperative bile duct injury within days to weeks of initial operation

usually present in one of two ways. One presentation is the progressive elevation of liver function test

results, particularly total bilirubin and alkaline phosphatase levels. These changes can often be seen as

early as the second or third postoperative day. The second mode of early presentation is with leakage of

bile from the injured bile duct. This presentation appears to occur most often in patients presenting with

bile duct injuries after laparoscopic cholecystectomy. Bilious drainage from operatively placed drains or

through the wound after cholecystectomy is abnormal and represents some form of biliary injury. In

patients without drains (including patients in whom the drains have been removed), the bile can leak

freely into the peritoneal cavity or it can loculate as a collection. Free accumulation of bile into the

peritoneal cavity results in either biliary ascites or bile peritonitis. Similarly, a loculated bile collection

can result in sterile biloma (Fig. 62-4) or in an infected subhepatic or subdiaphragmatic abscess.

Patients with postoperative bile duct strictures who present months to years after the initial operation

frequently have evidence of cholangitis. The episodes of cholangitis are often mild and respond to

antibiotic therapy. Repetitive episodes usually occur before the definitive diagnosis. Less commonly,

patients may present with painless jaundice and no evidence of sepsis. Finally, patients with markedly

delayed diagnoses may present with advanced biliary cirrhosis and its complications.

Figure 62-4. Large bile duct collection (biloma; arrow) occurring after bile duct injury. (Reproduced with permission from

Lillemoe KD, Pitt HA, Cameron JL. Postoperative bile duct strictures. Surg Clin North Am 1990;70:1355–1380.)

1609

Figure 62-5. A: Percutaneous transhepatic cholangiogram demonstrating bile duct stricture at hepatic duct bifurcation with

proximal duct dilatation. B: Percutaneous transhepatic cholangiogram demonstrating stricture (arrow) at a hepaticojejunostomy

anastomosis. BD, bile duct; int, intestine.

Laboratory Investigation

Liver function tests usually show evidence of cholestasis. In patients with bile leakage, the bilirubin can

be normal or minimally elevated because of absorption from the peritoneal cavity. When elevated,

serum bilirubin usually ranges from 2 to 6 mg/dL, unless secondary biliary cirrhosis has developed.

Serum alkaline phosphatase is usually elevated. Serum aminotransferase levels can be normal or

minimally elevated except during episodes of cholangitis. If advanced liver disease exists, hepatic

synthetic function can be impaired, with lowered serum albumin and a prolongation of prothrombin

time. Serum electrolytes and complete blood count are typically normal unless there is associated biliary

sepsis.

Radiologic Examination

The imaging techniques of abdominal ultrasound and computed tomography (CT) play an important

initial role in the evaluation of patients with benign postoperative biliary strictures. In patients who

present in the early postoperative period with evidence of a bile leak or biliary sepsis, these studies are

useful to rule out the presence of intra-abdominal collections that might require drainage (Fig. 62-4). CT

and ultrasound are also important in the initial evaluation of the patient presenting with a bile duct

stricture months to years after initial operation. Both studies can confirm biliary obstruction by

demonstrating a dilated biliary tree. CT is especially useful in identifying the level of obstruction of the

extrahepatic bile duct.

In patients suspected of having early postoperative bile duct injury, a radionucleotide biliary scan can

confirm bile leakage. In patients with postoperative external bile fistula, injection of water-soluble

contrast media through the drainage tract (sinography) can often define the site of leakage and the

anatomy of the biliary tree.

3 The “gold standard” for evaluation of patients with bile duct strictures is cholangiography.

Percutaneous transhepatic cholangiography (PTC) is usually more valuable than endoscopic retrograde

cholangiography (ERC) in patients with major bile duct injuries following laparoscopic

cholecystectomy. PTC is more useful in that as it defines the anatomy of the proximal biliary tree that is

to be used in the surgical reconstruction (Fig. 62-5). Furthermore, PTC can be followed by placement of

percutaneous transhepatic catheters, which can be useful in decompressing the biliary system both to

treat or prevent cholangitis and to control an ongoing bile leak. These catheters can also be of assistance

in surgical reconstruction and provide access to the biliary tree for nonoperative dilation. ERC is less

1610

useful than PTC in major bile duct transections during laparoscopic cholecystectomy because the

discontinuity of the extrahepatic bile duct usually prevents adequate filling of the proximal biliary tree

(Fig. 62-6). Often, ERC can demonstrate a normal-sized distal bile duct up to the site of the stricture

without visualization of the proximal biliary system (Fig. 62-7). This finding is frequently the case in

patients with injury during laparoscopic cholecystectomy, when the distal bile duct is often clipped and

divided. The development of magnetic resonance cholangiopancreatography (MRCP) has provided a

noninvasive technique that provides excellent delineation of the biliary anatomy. The quality of these

images has led some surgeons to advocate this technique as the initial step in the evaluation of patients

with suspected bile duct injuries and may eliminate the need for a diagnostic ERC in many patients.

MRCP is especially worth considering if the referral surgeon has a low pretest probability of utilizing

transhepatic stents for subsequent reconstructive purposes.

Figure 62-6. A: Endoscopic retrograde cholangiogram showing a relatively normal biliary tree in a patient with a postoperative

bile collection (see Fig. 62-5). B: Percutaneous transhepatic cholangiogram of the same patient, showing entire right hepatic

posterior lobe segment obstructed as the result of ligation of the segmental duct. The patient had an unrecognized anatomic variant

similar to that shown in Figure 62-2.

Preoperative Management

4 The preoperative management of a patient with a postoperative bile duct stricture depends primarily

on the timing of the presentation. Patients presenting in the early postoperative period can be septic

with either cholangitis or intra-abdominal bile collections. Sepsis must be controlled first with broadspectrum parenteral antibiotics, percutaneous biliary drainage, and percutaneous or operative drainage

of biliary leaks. Once sepsis is controlled, there is no hurry in proceeding with surgical reconstruction of

the bile duct stricture. The combination of proximal biliary decompression and external drainage allows

most biliary fistulas to be controlled or even to close. The patient can then be discharged home to allow

several weeks to elapse for resolution of the inflammation in the periportal region and recovery of

overall health.

1611