Key points

Introduction

The diagnosis of malignant biliary obstruction remains a significant clinical challenge. Accurately differentiating benign from malignant causes of a bile duct stricture is of obvious clinical importance for therapeutic planning and prognosis. The 2 most common causes of malignant strictures are cholangiocarcinoma (CCA) and pancreatic cancer. Diagnosis of these malignancies at an early stage can allow curative surgical resection or even liver transplantation for early-stage CCA. Tissue diagnosis of pancreaticobiliary malignancies via endoscopic approaches is well known to be limited by poor cellular yield and often requires surgical exploration for definite diagnosis. For cases of suspected pancreatic cancer, in which an extrinsic pancreatic mass is seen on cross-sectional imaging, or a double-duct sign (dilatation of both bile duct and pancreatic duct), endoscopic ultrasonography (EUS) should be the primary form of diagnosis and tissue should be sampled with fine-needle aspiration (FNA). This challenge in the evaluation of biliary strictures is made even more difficult in the case of an indeterminate stricture in which preprocedural cross-sectional imaging does not show an overt mass that would be highly suggestive of malignancy. Furthermore, surgical series show that 15% to 24% of patients who undergo resection for suspected malignant strictures based on preoperative imaging or endoscopic retrograde cholangiopancreatography (ERCP) ultimately have a benign diagnosis on pathology. This small but significant cohort of patients with benign strictures highlights the importance of accurate preoperative tissue diagnosis to avoid the morbidity and mortality of hepatobiliary surgery. This article reviews the causes of biliary strictures, the initial clinical evaluation of biliary obstruction, the diagnostic yield of ERCP-based sampling methods, the role of newer tools in the approach to evaluating strictures, and ways to address the ongoing challenge of stricture evaluation in patients with primary sclerosing cholangitis (PSC).

Introduction

The diagnosis of malignant biliary obstruction remains a significant clinical challenge. Accurately differentiating benign from malignant causes of a bile duct stricture is of obvious clinical importance for therapeutic planning and prognosis. The 2 most common causes of malignant strictures are cholangiocarcinoma (CCA) and pancreatic cancer. Diagnosis of these malignancies at an early stage can allow curative surgical resection or even liver transplantation for early-stage CCA. Tissue diagnosis of pancreaticobiliary malignancies via endoscopic approaches is well known to be limited by poor cellular yield and often requires surgical exploration for definite diagnosis. For cases of suspected pancreatic cancer, in which an extrinsic pancreatic mass is seen on cross-sectional imaging, or a double-duct sign (dilatation of both bile duct and pancreatic duct), endoscopic ultrasonography (EUS) should be the primary form of diagnosis and tissue should be sampled with fine-needle aspiration (FNA). This challenge in the evaluation of biliary strictures is made even more difficult in the case of an indeterminate stricture in which preprocedural cross-sectional imaging does not show an overt mass that would be highly suggestive of malignancy. Furthermore, surgical series show that 15% to 24% of patients who undergo resection for suspected malignant strictures based on preoperative imaging or endoscopic retrograde cholangiopancreatography (ERCP) ultimately have a benign diagnosis on pathology. This small but significant cohort of patients with benign strictures highlights the importance of accurate preoperative tissue diagnosis to avoid the morbidity and mortality of hepatobiliary surgery. This article reviews the causes of biliary strictures, the initial clinical evaluation of biliary obstruction, the diagnostic yield of ERCP-based sampling methods, the role of newer tools in the approach to evaluating strictures, and ways to address the ongoing challenge of stricture evaluation in patients with primary sclerosing cholangitis (PSC).

Causes of biliary strictures

The leading causes of malignant biliary obstruction are pancreatic cancer and CCA. Cholangiocarcinoma is a primary malignancy of the bile duct epithelium, and as such can involve both the intrahepatic and extrahepatic bile ducts. Worldwide, CCA accounts for 3% of all gastrointestinal malignancies and is the second most common primary liver malignancy after hepatocellular carcinoma. When CCA is diagnosed at an early T1 stage, surgical resection can have an excellent prognosis. The difficulty in the diagnosis of CCA is the poor cellular yield from the current first-line method of ERCP with brush cytology and/or biopsy. In addition, there is a known spectrum of benign causes of biliary strictures that can radiographically mimic CCA, making the exclusion of malignancy in these benign disorders clinically challenging.

Pancreatic cancer most often presents as a distal common bile duct stricture caused by extrinsic compression of the extrahepatic duct from a pancreatic head mass. This is in contrast with CCA, which often develops along the length of the bile duct, making its early detection particularly difficult because of the lack of a visible growth or tumor on imaging. Other less common malignant causes of biliary strictures include intraductal hepatocellular carcinoma, metastatic lesion, and extrinsic compression of the biliary tree from an associated visible mass or lymphadenopathy ( Table 1 ).

Benign biliary strictures can develop from a variety of causes ranging from recurrent cholangitis, iatrogenic causes (most commonly after cholecystectomy or liver transplantation), to cholangiopathy from autoimmune disease, human immunodeficiency virus, and PSC. One of the least understood mimickers of a malignant process is autoimmune or immunoglobulin G4 (IgG4)–associated sclerosing cholangitis (IgG4-SC). The prevalence and pathogenesis of this disease remains largely unknown but more than 80% of patients have increases of serum IgG4 levels to more than the upper limit of normal and a similar percentage of patients have an associated autoimmune pancreatitis. On cholangiogram, hilar IgG4-SC strictures are often indistinguishable from CCA; however, histology can be diagnostic, showing massive infiltration of IgG4-positive plasma cells with fibroinflammatory involvement of the submucosa of the bile duct wall.

Laboratory evaluation

The most common laboratory abnormality seen in patients with malignant biliary stricture is obstructive cholestasis. Direct hyperbilirubinemia is seen more commonly in patients with malignant obstruction than in those with a benign cause such as choledocholithiasis. Hyperbilirubinemia also has a higher likelihood of being associated with malignancy than increases in alkaline phosphatase level. Serum CA 19-9 (carbohydrate antigen 19-9) has been reported to have a sensitivity of 70% to 80% in the diagnosis of malignant strictures, with a specificity of 80% to 90%. The major limitation of CA 19-9 as a diagnostic tool of malignancy is its low specificity because its level can be increased in benign causes of cholestasis. In addition, the diagnostic accuracy of CA 19-9 in CCA is significantly lower than in pancreatic cancer, even when the cutoff values of CA 19-9 are increased. The interpretation of CA 19-9 as a diagnostic marker of bile duct cancer should be made in the context of the overall clinical impression, particularly when there is an acute cholangitis or cholestasis, which can cause false increases.

Imaging evaluation

MRI and magnetic resonance cholangiopancreatography (MRCP) have become essential parts of the baseline evaluation of patients presenting with biliary obstruction. MRCP has a very high sensitivity (96%–99%) for identifying the presence and anatomic location of a biliary obstruction but is primarily limited by poorer specificity (85%) in differentiating benign from malignant causes of obstruction. Cross-sectional imaging such as computed tomography (CT) can provide staging information such as nodal, vascular, and metastatic involvement in the presence of an overt mass but has poorer overall sensitivity for detection of CCA (only 40%–63%). Although no tissue diagnosis can be obtained from any imaging modality, they play an important role in providing the anatomic details needed for planning of ERCP-related interventions and preoperative staging before surgical resection.

Endoscopic retrograde cholangiopancreatography

The goals of endoscopic evaluation of a suspected malignant biliary stricture are to, first, obtain definite tissue diagnosis to obviate exploratory surgery and, second, to provide palliation of biliary obstruction with stent placement. The current first-line approach to endoscopic evaluation of biliary stricture remains ERCP with brush cytology and/or biliary biopsy. Although cholangiography provides clues to malignancy, such as complete obstruction, surface irregularity, location, and stricture length, these features cannot reliably distinguish the stricture’s cause and thus obtaining tissue for histopathology remains the gold standard. Routine brush cytology is known to have low sensitivity for diagnosing malignancy (23%–56%) despite a specificity of greater than 95%. There are a multitude of factors that contribute to the poor cytologic yield of ERCP, including tumor characteristics (submucosal growth pattern of CCA, malignant extrinsic compression), the anatomic location of the stricture (yield of biliary biopsy is higher in distal compared with proximal strictures), and the number of and processing of specimens. Several methods designed to improve the sensitivity of brush cytology have been studied with varying degrees of improvement. Fogel and colleagues reported on the use of a longer cytology brush with no significant improvement in the cancer detection rate. Attempts at stricture dilation with subsequent brush cytology sampling only marginally improved malignancy detection rate, from 27% to 34%. Endobiliary biopsy has a higher diagnostic yield compared with cytology alone ( Table 2 ), and when both modalities are combined sensitivity can be increased to 70% with preserved specificity of 100%. The addition of endoluminal FNA to the other 2 modalities, or triple tissue sampling, has the highest sensitivity at 77%. Wright and colleagues developed the so-called smash protocol for tissue sampling at ERCP to achieve a 72% on-site pathologic diagnosis for suspected biliary malignancy by changing the processing protocol for biopsies. The biliary biopsy tissue is smashed between 2 slides, stained by the Papanicolaou method with immediate on-site cytopathology review, and repeated sampling was obtained until a diagnosis was reached. This technique provides several distinct advantages, including its relative cost-effectiveness, on-site diagnosis, ease of performance, and low rate of complications despite multiple samplings, and it should be considered as a preferred method for biliary biopsy sampling in cases with high pretest suspicion for malignancy.

Fluorescence in situ hybridization

A promising tissue-based diagnostic tool that has recently been helping to improve the diagnostic yield of ERCP is fluorescence in situ hybridization (FISH) for detecting chromosomal aneuploidy or polysomy. These changes are thought to be present in an estimated 80% of pancreatobiliary malignancies. FISH uses fluorescently labeled DNA probes to detect cells with an abnormal number of chromosomes or mutations in a specific locus of a chromosome. There are 4 commercially available FISH probes, which bind to chromosomes 3 (CEP3), 7 (CEP7), 17 (CEP17), and the 9p21 locus of chromosome 9. The main advantage of FISH lies in its ease of use, because it can be performed on cells obtained from routine brush cytology samples during ERCP. The earliest prospective data on the diagnostic yield of FISH in indeterminate strictures by Levy and colleagues using CEP3, CEP7, and CEP 17 probes found that, in previously cytology-negative strictures, the sensitivity of FISH was 62% with specificity of 79% for malignancy. When trisomy 7 was considered a marker of malignancy in the combined cohort of patients both with and without PSC the overall sensitivity for diagnosing malignancy was 64%, specificity 82%, and diagnostic accuracy 72%. The main limitation of FISH is its reduced specificity in the setting of chronic inflammatory conditions such as PSC, in which polysomy can occur in the absence of CCA. Trisomy 7 in particular can be found in benign strictures of patients with PSC and decreases the specificity of FISH for malignancy in that challenging cohort. When a fourth FISH probe to the 9p21 locus of chromosome 9 (associated with mutation of the p16 tumor suppressor gene) is added to the repertoire the sensitivity of FISH can be improved significantly from 47% to 84% with preserved specificity of 97%. Other molecular-based techniques that have been examined to improve the diagnostic yield include bile aspirate analysis for p53 and KRAS mutations, but these are not currently considered part of the routine work-up and are still in the early phases of investigation.

Endoscopic ultrasonography: fine-needle aspiration

EUS with FNA has increasingly become part of the standard first-line evaluation of pancreatobiliary lesions as a complementary tool to ERCP for tissue sampling, tumor staging, and the exclusion of benign causes of biliary obstruction, such as stone disease. It has a reported overall sensitivity of between 43% and 86% for the diagnosis of all malignant strictures, but seems to be more sensitive in the evaluation of distal strictures, which are more often caused by pancreatic adenocarcinoma and more easily accessible for FNA sampling. EUS-FNA is generally less reliable in the evaluation of proximal CCA because its sensitivity decreases to 59% compared with 81% in distal CCA. The presence of a previously placed biliary stent can also decrease its sensitivity for malignancy detection because of a combination of stent-related acoustic shadowing, image degradation, and difficult needle access.

Perhaps the most concerning issue in the use of EUS-FNA in suspected CCA is the theoretic potential for malignant peritoneal seeding via the needle access pathway. This concern arises from a small number of case series of patients who underwent percutaneous biliary biopsies for CCA and developed carcinomatosis, with rates as alarmingly high as 83% in one series. Although there have been no cases reported of peritoneal carcinomatosis from EUS-FNA sampling of CCA, these concerns have led some transplant centers to consider tissue sampling via this method to be a contraindication for liver transplantation of hilar CCA. EUS-FNA remains an important complementary tool to ERCP in the evaluation of biliary strictures and can be particularly helpful in distal biliary strictures from pancreatic cancer, in cases of a suspected mass that may have been missed on other imaging techniques, or when ERCP-based tissue sampling techniques have been nondiagnostic.

Intraductal ultrasonography

Along with the evolution of EUS-FNA, intraductal ultrasonography (IDUS) was developed as a tool to enhance endobiliary imaging by the wire-guided placement of a high-frequency probe directly into the bile ducts. The main advantage of IDUS is its ease of use, because it can be done as part of routine ERCP without any need for biliary sphincterotomy or significantly prolonging the procedure. It can provide high-quality imaging of the periductal tissue along with limited tumor staging such as mass size and periportal vascular invasion (full lymph node staging still requires EUS). IDUS criteria for differentiating benign from malignant strictures have been established ( Box 1 ). Multiple studies show the sensitivity of IDUS to be 80% to 90%, specificity 83%, and it can improve the accuracy of ERCP from 58% to 83%. Note that IDUS ultimately does not provide histopathology but is another adjunctive tool to ERCP to help direct further work-up of these strictures.

Box 1

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

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related posts:

Difficult Biliary Access Endoscopic Retrograde Cholangiopancreatography for the Management of Common Bile Duct Stones and Gallstone Pancreatitis Endoscopic Retrograde Cholangiopancreatography in Surgically Altered Anatomy Endoscopic Treatment of Recurrent Acute Pancreatitis and Smoldering Acute Pancreatitis Sphincter of Oddi Dysfunction Innovations in Intraductal Endoscopy

Stay updated, free articles. Join our Telegram channel

Join