Bile Duct Cysts

Bile duct cysts typically are a surgical problem of infancy or childhood (Altman, 1994); however, in nearly 20% of patients, the diagnosis is delayed until adulthood. Although clinically similar, the presentation and therapeutic strategies for bile duct cysts in adults may differ substantially from those of younger patients. In contrast to the pediatric experience, adult patients have an increased rate of associated hepatobiliary pathology (Nagorney et al, 1984a; Powell et al, 1981; Rattner et al, 1983), and they often are first seen with complications of previous cyst-related procedures (Gigot et al, 1996; Nagorney et al, 1984a; Ono et al, 1982; Todani et al, 1984a). The surgical management of bile duct cysts in adults is thus complicated by coexisting hepatobiliary disease and the added technical difficulties of reoperative biliary surgery. Despite the heterogeneity of the disease and the absence of clinical trials, a consensus on the management of extrahepatic bile duct cysts has been established: excision is best. However, the management of intrahepatic bile duct cysts remains problematic, and the method of choice for reestablishing bilioenteric continuity after excision is debatable. This chapter examines the spectrum of hepatobiliary pathology encountered in adults with bile duct cysts and describes the alternative surgical approaches for managing such patients.

Diagnosis

Classification

Bile duct cysts are classified on the basis of site, extent, and shape of the cystic anomaly of the ductal system. Although the term choledochal cyst is often used, bile duct cyst is semantically more appropriate because cystic dilation can occur anywhere throughout the biliary ductal system, not just in the common bile duct (choledochus). The first coherent classification of extrahepatic bile duct cysts was proposed by Alonso-Lej and colleagues in 1959. Cysts were classified into three types: type I is a fusiform or saccular dilation of the common hepatic and common bile duct, type II is a supraduodenal diverticulum of the common hepatic or common bile duct, and type III is an intraduodenal diverticulum of the distal common bile duct or choledochocele. Although this classification did not include intrahepatic bile duct cysts, this simple and practical scheme has provided the basis for all other classification systems.

The recognition of intrahepatic bile duct cysts prompted modification of the Alonso-Lej classification system, and in 1958, Caroli and colleagues described the entity of multiple intrahepatic bile duct cysts in the absence and presence of extrahepatic cysts. Although initially described as an entity of multiple saccular dilations of only the intrahepatic ducts, the term Caroli disease has been broadly applied to describe any patient with intrahepatic bile duct cysts, regardless of the presence of extrahepatic bile duct cysts or the shape of the intrahepatic cysts (Dayton et al, 1983). Caroli disease represents a spectrum of diseases that may include such variants as cysts with congenital hepatic fibrosis or Grumbach disease (see Chapter 70A; Grumbach et al, 1954). Todani refined the classification system of bile duct cysts by combining the Alonso-Lej classification and the variants of Caroli disease (Todani et al, 1977). The comprehensive Todani classification system is shown in Fig. 46.1.

FIGURE 46.1 Classification of bile duct cysts.

Matsumoto also modified the Alonso-Lej classification system based not only on the location of the cyst but also on the configuration of cysts (Matsumoto et al, 1977a). To date, clinical management of bile duct cysts is dictated by cyst location, not configuration; therefore there is little clinical rationale for adoption of the more detailed classification scheme (Kamisawa et al, 2005; Komi et al, 1992; Matsumoto et al, 1977b), although further additions to the Todani classification have been suggested (Calvo-Ponce et al, 2008; Conway et al, 2009).

Etiology

Multiple theories have been proposed to explain the origin of bile duct cysts. The most widely accepted hypothesis is that cystic dilation of the bile ducts is related to an anomalous arrangement of the pancreaticobiliary ductal junction (Babbitt, 1969; Cheng et al, 2004; Komi et al, 1992; Okada et al, 1990; Todani et al, 1984b; Wiedmeyer et al, 1989). The pancreaticobiliary junction is proximal to the sphincter of Oddi (Fig. 46.2) (Nagata et al, 1986). An anomalous pancreaticobiliary junction is often associated with a long common channel that predisposes to reflux of pancreatic juice into the biliary tree, leading to inflammation, ectasia, and, ultimately, to dilation. This postulate has been supported by biliary manometric studies (Iwai et al, 1986), high concentrations of pancreatic enzymes in cyst fluid (Todani et al, 1990), and histopathologic studies of ductal epithelial hyperplasia, round cell infiltration, and marked ductal fibrosis (Oguchi et al, 1988). Moreover, experimental canine studies of pancreaticocholecystostomy and pancreaticocholedochostomy have resulted in cystic dilation of the extrahepatic bile duct (Oguchi et al, 1988; Ohkawa et al, 1981). Pancreaticobiliary maljunction, or anomalous union, defined as union of the pancreatic and biliary duct outside of the duodenal wall, has also been implicated in biliary tract cancer even in the absence of a biliary cyst (Funabiki et al, 2009; Horaguchi et al, 2005).

FIGURE 46.2 A long common channel, as shown in this figure of the pancreaticobiliary ductal confluence, is the most accepted cause of choledochal cysts.

Reports of choledochal cysts of the same type in family members suggest that a hereditary factor may contribute, albeit rarely, to bile duct cysts (Iwafuchi et al, 1990). Finally, oligoganglionosis in the distal neck of the cyst may account for some bile duct cysts. The reduction in ganglion cells in the narrow portion of the cyst wall may be the biliary equivalent of Hirschsprung disease of the colon (Kusunoki et al, 1988).

Demographics

Bile duct cysts are an uncommon finding, with fewer than 4000 cases reported (Flanigan, 1975; Powell et al, 1981; Yamaguchi, 1980). Estimates of the actual clinical incidence range from one in 13,000 to one in 2 million patients (Olbourne, 1975), and biliary cysts account for approximately 1% of all benign biliary disease (Saxena et al, 1988). The estimated prevalence of bile duct cysts by type (Todani classification), based on a review of the current literature, is shown in Table 46.1. The true prevalence of bile duct cysts by current classification schemes is underestimated here because classification is often not detailed. The frequency of bile duct cyst in decreasing order is the type I choledochal cyst (79%), followed by type IV (13%), then type III, or choledochocele (4%), and type II (2.6%). Multiple intrahepatic bile duct cysts without an extrahepatic component (Caroli disease) occur in fewer than 1% of all patients with biliary cystic disease. The distribution of cyst types encountered in adults is similar to that in infants and children (Gigot et al, 1996; Nagorney et al, 1984a; Ono et al, 1982), except type IV cysts are more frequent in adults (Todani et al, 1978, 1984a).

A marked population prevalence has been seen in Japan (Flanigan, 1975; Powell et al, 1981; Yamaguchi, 1980), where more than one third of reported cases have occurred. Although the number of reported cases has increased recently, this finding probably reflects advances in diagnosis through improvements in hepatobiliary imaging rather than a true increase in incidence.

A female preponderance among patients with bile duct cysts is well known, regardless of cyst type. In Flanigan’s (1975) collected series of 820 cases, 81% of the patients were female. A similar female/male ratio has been found in adults (Powell et al, 1981). Current theories on the pathogenesis of bile duct cysts have not implicated sex hormones or congenital intrauterine endocrine disturbances as possible factors.

Clinical Features

Choledochal cysts may remain asymptomatic for many years. Initial clinical presentation in adulthood (age >16 years) occurs in fewer than 20% of all patients with common duct cysts (Flanigan, 1975). Diagnosis may be an incidental finding on imaging studies for unrelated processes. If symptomatic, bile duct cysts usually present with symptoms mimicking calculous biliary tract disease, regardless of cyst type. Symptoms are typically intermittent, recurrent epigastric or right hypochondrial pain, abdominal tenderness, fever, and mild jaundice—the most common initial findings. The pain may radiate to the right infrascapular region or to the mid back, and it generally persists for hours. Abdominal pain or discomfort is often overshadowed by fever and rigors, which may occur repeatedly for several days. An abdominal mass is uncommon in adults; however, if a mass is present, cyst-associated malignancy must be strongly suspected. Biliuria heralds the onset of clinical jaundice. Nausea, vomiting, and anorexia may accompany other symptoms; if cholangitis persists, the jaundice deepens and overt signs of sepsis may evolve.

Approximately 15% of adults with common duct cysts present with overt evidence of cirrhosis or hepatic fibrosis from chronic biliary obstruction. Such patients typically have had multiple operations for complications of type I or IVA cysts or Caroli disease. Hepatomegaly and splenomegaly are common in patients with cirrhosis and portal hypertension. Hematemesis, melena, and ascites may accompany portal hypertension; however, cholangitis remains the most common initial symptom complex in adults, whether or not cirrhosis is present. Interestingly, other signs of chronic liver disease—muscle wasting, fatigue, spider angiomas, and pruritus—are uncommon. Liver failure is seen late in Caroli disease.

Clinical pancreatitis is present in nearly 30% of patients with choledochal cysts (Nagorney et al, 1984a). In contrast to patients with cholangitis, patients with pancreatitis have more intense and prolonged epigastric pain and vomiting. Fever and jaundice are less prominent. Weight loss, although unusual, is noteworthy because nearly 70% of adults with this finding will harbor an associated bile duct malignancy (see Chapter 50B).

Imaging

Accurate preoperative diagnosis can be achieved by using the routine diagnostic armamentarium for patients with suspected biliary tract disease: abdominal ultrasonography (US), computed tomography (CT), percutaneous transhepatic cholangiography (PTC), endoscopic retrograde cholangiopancreatography (ERCP), and magnetic resonance cholangiography (MRC) (Fig. 46.3). Current reviews of the diagnostic imaging modalities of bile duct cysts with representative images are referenced (Kim et al, 1995; Savader et al, 1991a, 1991b). Although hepatobiliary scintigraphy has also been useful in identifying bile duct cysts, its value is limited because the information is functional and not anatomic; it is therefore only complementary to US, CT, and direct cholangiography. In general, bile duct cysts are typically recognized serendipitously in adults, unless a past diagnosis had been established before adulthood.

FIGURE 46.3 Ultrasonography (A), endoscopic cholangiography (B), and computed tomography (C) showing a type IVA cyst with fusiform dilation of the common bile duct and the left intrahepatic ductal system.

The noninvasiveness and accuracy of US, combined with its ability to image adjacent viscera, support its use as the initial investigative procedure (see Chapter 13). Real-time US is favored because of its speed and ease of application. Moreover, the ability of real-time US to scan multiple planes adds a distinct advantage in defining the extent of the dilation. The sonographic features of choledochal cysts have been well defined for type I choledochal cysts and the variants of Caroli disease (Bass et al, 1977; Morgan et al, 1980; Todani et al, 1978; Young et al, 1990). US of type I cysts simply shows an irregular hypoechoic segmental dilation of the extrahepatic bile duct. Focal duct wall thickening or nodularity should arouse suspicion for cancer in an adult. Stones within the cyst can also be identified by typical echogenic features and acoustic shadowing. The absence of septations on US will distinguish choledochal cysts from extrahepatic biliary tumors, such as cystadenomas (Nagorney et al, 1984b). Although the sensitivity of US is excellent for cysts involving the bile duct proximal to the pancreas, US is limited in adults in identifying choledochoceles because of the small size of these cysts and the frequency of bowel gas overlying the terminal common bile duct.

Caroli disease is clinically recognized by multiple cysts adjacent to the major intrahepatic bile ducts on US (Bruneton et al, 1983). Cysts may be unilobar or bilobar, and confirmation of communication of the intrahepatic cysts to the bile ducts can be confirmed by scintigraphy (Sty et al, 1978) or using CT with biliary contrast medium (Musante et al, 1982).

CT combined with intravenous cholangiography is useful for demonstration of cyst communication with the biliary tract (see Chapter 16; Hoglund et al, 1990). Intravenous cholangiography is performed 2 hours before abdominal CT scanning. The sensitivity of CT allows for accurate identification of accumulated contrast material within the cyst, if communication is present, and for accurate definition of the bile duct cyst (Tohma et al, 2009).

Direct cholangiography is the preferred diagnostic modality for accurate definition of the type of choledochal cyst (see Chapter 18). Indeed, cyst classification is based on cholangiographic features (Matsumoto et al, 1977a; Todani et al, 1977). All bile duct cysts communicate with the ductal system. Direct cholangiography is a prerequisite to operation because it has the advantage of being able to define the configuration and extent of the cyst; it also can identify ductal strictures, stones within the biliary and pancreatic ducts, and polypoid filling defects that suggest ductal malignancy. In addition, direct cholangiography is essential in defining the relationship of the distal bile duct in the type I cyst or the extrahepatic component of type IVA cyst to the pancreatic duct. The base of the extrahepatic choledochal cyst typically joins the pancreatic duct 2 to 4 cm proximal to the duodenum (Fig. 46.4), resulting in a long common channel (ampulla) (Jona et al, 1979; Komi et al, 1992; Ono et al, 1982; Rattner et al, 1983). The angle of fusion between the distal bile duct and pancreatic duct varies widely and has led to subclassifications of cyst type (Komi et al, 1992). Anatomic definition of the pancreatic biliary ductal junction is of key importance to avoid intraoperative damage of the pancreatic duct during cyst excision, recognize stones impacted within the common channel or junction, and exclude distal tumors. Regardless of method, complete cholangiographic visualization of the entire biliary and pancreatic ductal systems is essential in patients with choledochal cysts because failure to recognize pancreatic duct anomalies or segmental areas of dilation within the liver parenchyma may lead to sepsis, subsequent cholangitis, pain, pancreatitis, and eventual reoperation.

FIGURE 46.4 Endoscopic retrograde cholangiopancreatogram shows the most frequent anomaly of the bile duct, cyst, and pancreatic duct anatomy with perpendicular fusion of the distal bile duct or cyst to the pancreatic duct and the presence of a long common channel.

There are advantages and disadvantages of PTC and ERCP for assessing choledochal cysts. Regardless of method, large volumes of radiographic contrast may be required for complete visualization of the bile ducts. Adults without previous cystenterostomy are probably best evaluated by ERCP because it provides better visualization of an abnormal pancreaticobiliary ductal junction (Savader et al, 1991b; Komi et al, 1992). Carcinoma can be excluded by biopsy or brush cytology, and intracystic stones can be extracted after papillotomy to relieve severe cholangitis before surgery. Endoscopy also allows visualization of the esophagus and stomach to exclude signs of portal hypertension. The endoscopist should attempt to examine the ductal bifurcation and the lining of the cyst, if a patent cystoduodenostomy allows introduction of the endoscope into the biliary tree. Endoscopically directed biopsy of intracystic masses should be performed to exclude malignancy. Obstructing balloons should be available to ensure that complete radiographic filling of the biliary tree is possible, especially in patients with prior cystoduodenostomy. Endoscopy of the cyst through a cystenterostomy may also permit diagnosis of intraductal stenoses by membranes or septa at the confluence of the major bile ducts in patients with type IVA cysts (Ando et al, 1995). The procedure of choice for type III cysts or choledochoceles is ERCP because endoscopic papillotomy is potentially therapeutic (Sarris & Tsang, 1989; Venu et al, 1984).

PTC is an effective alternative to ERCP in the diagnosis of choledochal cysts. Because of endoscopic limitations imposed by the Roux-en-Y jejunal limb, PTC is particularly advantageous in patients with previous Roux-en-Y cystenterostomy and in patients with type IV bile duct cysts, in whom ductal strictures or tumors prevent complete visualization of intrahepatic cysts by ERCP (Savader et al, 1991b). The major diagnostic disadvantage of PTC is occasional failure to clearly define the pancreaticobiliary junction. Percutaneous biliary drainage (PBD) may be performed after PTC when indicated for control of biliary sepsis or to aid surgical reconstruction. PTC is particularly limited in patients with a widely patent cystenterostomy or huge extrahepatic cysts, because intracystic contrast is often superimposed over the pancreaticobiliary junction and obscures its radiographic definition.

MRC provides a noninvasive method of imaging the bile ducts and its anomalies (see Chapter 17), and it has been shown to provide an accurate anatomic definition of bile duct cysts in neonates, infants, and young children (Fitoz et al, 2007; Miyazaki et al, 1998). In a recent study, MRC reliably diagnosed biliary cystic disease in 74 (96%) of 76 patients with an accuracy of 86% for ductal anomalies (Park et al, 2005); however, it was less effective in detecting minor ductal anomalies and small cysts (Shaffer, 2006). MRC application to adults can be anticipated because image acquisition in adults is not hindered by the technical difficulties met when performing these studies in children. The limitation of MRC for bile duct cysts has related to its inability to clearly define the pancreaticobiliary junction (diagnostic accuracy, 69%) and its lack of therapeutic capability.

Associated Hepatobiliary Pathology

Additional hepatobiliary pathology is frequently associated with bile duct cysts in adults. Cystolithiasis, hepatolithiasis, calculous cholecystitis, pancreatitis, cholangiocarcinoma, intrahepatic abscess, and cirrhosis with portal hypertension are potential problems that may either precipitate or complicate treatment. Although spontaneous perforation of bile duct cysts has been reported in infants and children, that complication has not been recognized in adults (Ando et al, 1998). Studies in adults suggest that nearly 80% of adults with bile duct cysts have complications from one or more of the conditions listed above (Nagorney et al, 1984a; Ono et al, 1982). A recent study suggests that complications in adults may decrease if excision is used routinely in infants and children (Gigot et al, 1996).

Cystolithiasis is the most frequent accompanying condition in adults with bile duct cysts. In contrast to the low prevalence of cystolithiasis in pediatric patients (Flanigan, 1975; Matsumoto et al, 1977a, 1977b; Rattner et al, 1983), the prevalence of intracystic stones has ranged from 2% to 72% in adults (Chijiiwa & Koga, 1993; Nagorney et al, 1984a; Todani et al, 1988). Although their composition has not been analyzed biochemically, most intracystic stones have been described as soft, earthy, and pigmented in appearance, thus supporting bile stasis as a primary etiologic factor (Matsumoto et al, 1977a, 1977b). These stones typically are associated with thick, viscous bile that may form bile duct or cyst casts. Cystolithiasis frequently complicates anastomotic strictures after previous cystoenterostomies, which suggests that stasis and infection are major factors contributing to the pathogenesis of these stones.

Hepatolithiasis has been recognized with increasing frequency with long-term follow-up and may occur with or without evidence of anastomotic stricture (Fig. 46.5) (see Chapter 7; Deziel et al, 1986; Gigot et al, 1996; Uno et al, 1996). Some patients will have a complete or partial stricture of the cystenteric anastomosis, and hepatolithiasis develops as a consequence of proximal bile stasis or migration of intracystic stones. Hepatolithiasis usually occurs in type IV bile duct cysts. A recent study has shown that more than 80% of type IV bile duct cysts are associated with either a membranous or septal stenosis of the major lobar bile ducts near the confluence (Ando et al, 1995). Stenosis of the major ducts should be assessed in all patients with hepatolithiasis. Intrahepatic stones may be sequestered in segmental ducts, leading to further localized intrahepatic ductal dilation or subsequent intrahepatic abscess formation.

FIGURE 46.5 Endoscopic retrograde cholangiogram of a type IVA bile duct cyst showing multiple intrahepatic stones and a patent choledochoduodenostomy.

Gallbladder disease is a frequent complication for patients with bile duct cysts. Gallbladder disease can occur de novo, leading to the incidental discovery of bile duct cysts, or it can evolve after treatment for bile duct cysts. In particular, a gallbladder that is not excised as part of the initial treatment for symptomatic bile duct cysts is frequently a subsequent source of biliary tract disease. Both acute and chronic cholecystitis, with or without stones, has been recognized in these patients. Similar to stones within cysts, stones in noncystic portions of the bile duct are typically bilirubinate in appearance. Stasis is probably the primary etiologic factor in such cases.

The association of pancreatitis with choledochal cysts is a well-recognized entity, particularly in adults (see Chapter 53). The prevalence has ranged from 2% to 70% (Lipsett et al, 1994; Nagorney et al, 1984a; Okada et al, 1990; Rattner et al, 1983; Swisher et al, 1994). The true incidence of pancreatitis is difficult to determine because the diagnosis is based on the clinical findings of epigastric pain, nausea and vomiting, and hyperamylasemia. Another clinical condition, the common channel syndrome (Jona et al, 1979) or pseudopancreatitis (Todani et al, 1990), may mimic these findings completely, thus overestimating the true incidence of pancreatitis. Whether differentiation between these two entities has clinical or therapeutic significance is unclear. Pancreatitis associated with bile duct cysts typically is mild. The pattern of pancreatitis is acute and often relapsing, but objective pancreatic dysfunction seldom occurs. Chronic pancreatitis associated with bile duct cysts is rare.