CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2009;7:S79 S83 Differential Diagnosis and Treatment of Biliary Strictures KAZUO INUI, JUNJI YOSHINO, and HIRONAO MIYOSHI Department of Internal Medicine, Second Teaching Hospital, Fujita Health University, Nagoya, Japan Biliary tract strictures present both diagnostic and therapeutic challenges to clinicians. Advances in imaging and endoscopic techniques have improved our ability to differentiate between benign and malignant lesions. Intraductal ultrasonography (IDUS), using an endoscopic approach, has the potential to aid in separating benign and malignant biliary lesions. In a series of 93 patients, a majority of whom had cancer, we found that IDUS had a sensitivity and specificity of 89.7% and 84%, respectively, for diagnosing biliary strictures. However, benign strictures associated with untreated autoimmune pancreatitis and/or the intrapancreatic portion of the distal common bile duct could not be easily distinguished from malignant strictures. Direct visualization of biliary mucosa using a percutaneous transhepatic endoscopic approach also helps separate benign from malignant biliary strictures. Further, the ability to obtain multiple directed biopsies using a percutaneous approach also increases diagnostic accuracy. A final advantage of the percutaneous approach is that once a suitable sized tract has been established, biliary strictures and stents can be placed. IDUS and percutaneous biliary endoscopy are promising new modalities for the diagnosis and treatment of biliary strictures. Diagnosis of tumor extension in biliary carcinoma has been made more comprehensive and reliable by advances in diagnostic imaging including multiple-row detector computed tomography and magnetic resonance cholangiopancreatography. However, the differential diagnosis between malignant and benign biliary strictures still is difficult, because multiple-row detector computed tomography and magnetic resonance cholangiopancreatography have limited abilities to delineate small lesions. 1 Our group has used intraductal ultrasonography (IDUS) 2,3 and percutaneous transhepatic cholangioscopy (PTCS) 4,5 for detailed assessment of biliary strictures. The following describes the usefulness of these procedures as well the use of a percutaneous transhepatic treatment approach for benign biliary strictures. Differential Diagnosis Intraductal Ultrasonography Instruments and methods. We usually used a Ropeway System probe, 2.5 mm in diameter (UM-DG-35R; Olympus Medical Systems, Tokyo, Japan), incorporating a radial scanning system with a frequency of 20 MHz. 6 The probe was connected to an endoscopic ultrasonic observation unit (EU- M2000; Olympus Medical Systems) and controlled by a probedriving unit (MAJ 2000; Olympus Medical Systems). For 3-dimensional IDUS, we used an ultrasound image processing unit (EU-IP2; Olympus Medical Systems) to produce reconstruction images such as dual-plane reconstructions. The probe was passed easily through the 2.8 mm diameter biopsy channel of an electronic duodenoscope (JF 260V or TJF 200; Olympus Medical Systems). We performed IDUS by either a transpapillary or a percutaneous transhepatic approach: the area surveyed by the probe was confirmed by fluoroscopy. Patients. From 1995 to 2006, we performed IDUS in 93 patients with biliary strictures: 38 patients had bile duct carcinoma; 23, pancreatic carcinoma; 7, gallbladder carcinoma; 9, benign stricture; 6, chronic pancreatitis; 6, autoimmune pancreatitis; and 4, sclerosing cholangitis. In 42 patients with a malignant stricture, a definitive diagnosis was made by histologic examination of the surgical specimen; in 26 patients with malignant strictures and 9 with benign strictures, the diagnosis was made by histologic examination of biopsy specimens. Findings by IDUS. The normal bile duct wall is visualized as having 2 to 3 layers. In contrast, IDUS shows carcinoma as a hypoechoic area of irregular thickness. In the upper and middle portions of the extrahepatic bile duct, benign strictures show regular thickening of the bile duct wall by IDUS. In malignant strictures, IDUS showed irregular thickening of the wall, with either a smooth outer hyperechoic layer or an irregular outer hyperechoic layer. In the lower portion of the extrahepatic bile duct, benign strictures were not associated with a mass, but with coarse echoes in the parenchyma of the pancreas. In malignant strictures, IDUS demonstrated an irregular hypoechoic mass related to the bile duct. Results. When wall thickening was absent or regular, the stricture was benign (Figure 1A). When wall thickening was irregular, the stricture was malignant (Figure 1B). Irregularity of the outer hyperechoic layer indicated invasion of deep tissues (Figure 1C). Sometimes, difficulty arose in correctly diagnosing benign strictures in patients with autoimmune pancreatitis, where IDUS displayed irregular wall thickening in the acute phase. However, after steroid therapy, wall thickness decreased conspicuously in these patients. Another characteristic IDUS finding in patients with autoimmune pancreatitis was wall thickening in dilated or nonstenotic bile ducts. In 8 of 93 patients overall 5 of 23 with pancreatic cancer and 3 of 6 with autoimmune pancreatitis IDUS could not distinguish the bile duct wall from the pancreatic parenchyma (Figure 2). In patients with biliary strictures in the lower portion of the extrahepatic bile duct, marked echo attenuation confounded the Abbreviations used in this paper: IDUS, intraductal ultrasonography; NBI, narrow band imaging; PTBD, percutaneous transhepatic biliary drainage; PTCS, percutaneous transhepatic cholangioscopy. 2009 by the AGA Institute 1542-3565/09/$36.00 doi:10.1016/j.cgh.2009.08.027
S80 INUI ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 11S Figure 1. (A) IDUS shows regular thickening of the bile duct wall in a patient with benign stricture. (B) IDUS shows irregular thickening (white arrows) of the bile duct wall in a patient with cholangiocarcinoma. (C) IDUS shows irregularity of the outer hyperechoic layer (white arrows) in a patient with cholangiocarcinoma invading to deep tissues. differential diagnosis. The overall accuracy of IDUS in differential diagnosis of biliary strictures was 88.2%; sensitivity, 89.7%; and specificity, 84%. Thus, IDUS was useful for differentiating between malignant and benign biliary strictures (Figure 3). Percutaneous Transhepatic Cholangioscopy Instruments and Methods To perform PTCS, we used a cholangioscope developed by Olympus Medical Systems (CHF-P20). The instrument, 4.9 mm in outer diameter with a 2.8 mm working diameter was inserted into the bile duct through a sinus tract created for percutaneous transhepatic biliary drainage (PTBD). We performed PTCS 7 to 10 days after the initial PTBD. After administration of atropine sulfate (0.5 mg intramuscular) and diazepam (10 mg intravenous) and insertion of a guidewire into the sinus tract under fluoroscopy, the drainage tube was removed. The cholangioscope was inserted into the biliary tract along with the guidewire. Patients From 1989 to 2000, we performed PTCS in 317 patients with biliary strictures. Here, we report on 28 patients with cholangiocarcinoma and 16 with benign strictures (9 from cholangitis and 7 from chronic pancreatitis). Findings by PTCS Endoscopic findings by PTCS in cholangiocarcinoma included papillary or nodular protruding lesions, tumor vessels, and irregular, granular mucosa. Papillary protrusion was observed in 9 of 28 cancer patients (32%) and nodular protrusion in 15 of 28 (54%). These findings, suggestive of cholangiocarcinoma, were not seen in patients with benign biliary strictures. Tumor vessels (Figure 4) were observed in 26 of 28 patients (93%). Irregular, granular mucosa was observed in 11 of 28 Figure 2. IDUS showing marked echo attenuation in a patient with chronic pancreatitis. IDUS could not distinguish the bile duct wall from the pancreatic parenchyma. Figure 3. Decision tree of biliary strictures with IDUS.
November 2009 DIAGNOSIS AND TREATMENT OF BILIARY STRICTURES S81 without prominent vessels. The light uniform color of the mucosa shown by NBI allowed tumor vessels to be sharply highlighted. Treatment Instruments and Methods We performed dilation of strictures through the sinus tract made for PTBD. After a diagnosis of benign stricture was made based on PTCS findings and pathologic examination, a 10 F catheter was inserted along a guidewire to the distal common bile duct or into the duodenum. The size of the tract was gradually enlarged using successively larger catheters passed over a guidewire in 1 day. Finally, a 16 F catheter was inserted and left in place. After placement of the catheter and several weeks, we placed the catheter above the stricture for 1 or 2 days. Subsequently, if no recurrence of stricture was seen without a catheter, the catheter was removed completely. Figure 4. PTCS displayed tortuous vessels at the stenosis of the bile duct which we diagnosed as tumor vessels. patients (39%) and suggested superficial extension of the cholangiocarcinoma. Results After careful characterization of the stenosis, we performed biopsy under endoscopic observation. Biopsy was easier to perform by PTCS than by a transpapillary approach. Although pathologic examinations of biopsied specimens obtained by PTCS were the most reliable method for diagnosing cancer, the macroscopic PTCS findings still were helpful. One of the most important endoscopic findings was greater mucosal vascularity in malignant than in benign biliary strictures. Recent advances now allow observation with narrow band imaging (NBI), which shows mucosal details, including vascular structure, more clearly than regular illumination (Figure 5). With NBI, the normal bile duct mucosa appeared whitish, Patients We performed percutaneous dilation of a benign biliary stricture in 6 patients (3 men, 3 women). The patients ranged in age from 52 to 88 years (mean, 74.5 years). Diagnosis was made by histologic examination of biopsy specimens. Results All 6 patients had successful dilation by the method above, and have not required further biliary stenting during a mean observation period of 4 years. Duration of stenting by the PTBD catheter ranged from 8 to 178 days (mean, 77 days). The observation period after catheter removal ranged from 8.1 to 60.4 months. No patients have needed further PTBD catheter treatment. Case 1 A 77-year-old woman was admitted for evaluation and treatment of liver dysfunction. Ultrasonography depicted a hypoechoic mass in the upper portion of the extrahepatic bile duct. Multiplanar reconstruction imaging by contrast-enhanced helical computerized tomography showed an enhanced lesion in this area. A cholangiogram performed via a PTBD Figure 5. With (A) NBI, PTCS depicted the mucosa more clearly than (B) ordinary observation.
S82 INUI ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 11S Figure 6. (A) A cholangiogram performed via PTBD of sinus tract depicted obstruction of the proximal extrahepatic bile duct. (B) A 16 F catheter was inserted into the stricture and kept inserted as a stent for 8 days. (C) The stenosis was improved after removal of the catheter. sinus tract showed obstruction of the proximal extrahepatic bile duct (Figure 6A). PTCS demonstrated an irregular stenosis of the bile duct, but showed no tumor vessels. Instead, the lesion looked like an ulcer scar. We obtained several biopsy specimens, which showed benign inflammation of the bile duct. We carried out gradual dilation of the stricture via the PTBD sinus tract using progressively larger catheters passed along a guidewire. Finally, a 16 F catheter was inserted (Figure 6B). The 16 F catheter remained in place as a stent for 8 days to dilate the stricture (Figure 6C). After catheter removal, the patient has remained free from symptoms and liver dysfunction for 11 years. Case 2 A 60-year-old man was admitted to our hospital because of jaundice. PTBD and dilation using the sinus tract were performed on the same day. PTBD cholangiography indicated a regular stenosis of the distal common bile duct (Figure 7A). We obtained several biopsy specimens, in which the pathologic findings indicated benign disease. A 10 F catheter was passed along a guidewire to the duodenum, with subsequent gradual substitution of larger catheters up to 16 F (Figure 7B). The 16 F catheter could not be removed for 6 months because of recurrent stenosis. PTBD cholangiography 6 months after dilation with a bougie confirmed dilation of the stricture (Figure 7C). No recurrence has been seen over 9 years. Discussion Biliary tract IDUS has been reported to be useful in diagnosis of choledocholithiasis, 7 differential diagnosis of biliary strictures, 8 and the local staging of cholangiocarcinoma. Many authors 9 12 have emphasized that IDUS is reliable in evaluation of cholangiocarcinoma. Additionally, we reported the usefulness of 3-dimensional IDUS as a comprehensive image display in 1997. 13 Menzel et al 8 reported that a malignant tumor should be suspected when hypoechoic masses show irregular margins and inhomogeneous hypoechoic areas invading the surrounding tissue. Penetration was defined as continuation of the main hypoechoic tumor mass into adjacent structures. On the other hand, homogeneous echo patterns and smoothly defined margins argued against malignant disease. They reported that ac- Figure 7. (A) PTBD cholangiography indicated a regular stenosis of the distal common bile duct. (B) A 10 F catheter was passed along a guidewire to the duodenum, with subsequent gradual substitution of larger catheter up to 16 F. (C) PTBD cholangiography 6 months after dilation with a bougie confirmed dilation of the stricture.
November 2009 DIAGNOSIS AND TREATMENT OF BILIARY STRICTURES S83 curacy, sensitivity, and specificity of IDUS, respectively, were 89.1%, 91.1%, and 80% for differentiating between benign and malignant disease. When we focused on regularity of bile duct wall thickness, IDUS accuracy was 88.2%; sensitivity, 89.7%; and specificity, 84% for differential diagnosis between benign and malignant disease. However, in this series, we encountered difficulty in diagnosing the etiology of stenosis in the lower portion of the extrahepatic bile duct, because patients with chronic pancreatitis or autoimmune pancreatitis showed marked echo attenuation surrounding the ultrasonic probe, a pattern also seen in cancer. Recently, sclerosing cholangitis occurring with autoimmune pancreatitis has been noted widely. Findings by ultrasonography or other imaging modalities have been reported, but not those by IDUS. Benign stricture is difficult to diagnose in patients with autoimmune pancreatitis, because IDUS shows irregular wall thickening in the acute phase. Decreased wall thickness after steroid therapy is helpful in diagnosis of autoimmune pancreatitis, as are IDUS findings of wall thickening in a dilated or nonstenotic bile duct. PTCS is reliable for precise diagnosis of the biliary strictures, 14,15 while pathologic examination of specimens obtained by PTCS were most reliable. PTBD is required before PTCS, but biopsy by PTCS is easier than via a transpapillary approach. Recent technical advances permit enhanced observation of the mucosa using NBI. Using NBI, the bile duct mucosa appeared whitish, without prominent vessels. Against such a background, any tumor vessels present were easily visualized. Knowledge of the PTCS findings in cholangiocarcinoma should be useful in peroral cholangioscopy. We performed dilation of strictures through the sinus tract used for percutaneous transhepatic biliary drainage. All patients strictures were dilated successfully by this method, and patients have not required further biliary stenting interventions during a mean observation period of 4 years. However, the required duration of stenting by the catheter ranged from 8 to 178 days (mean, 78 days). In half of the patients, the duration of stenting was less than 1 month; in the others, stenting was required for about 6 months. This variability may reflect differences in severity of fibrosis in surrounding tissues, especially when the stricture occurs in the intrapancreatic portion of the common bile duct. In patients who have clearly defined benign biliary strictures, endoscopic biliary stenting by a transpapillary approach is the preferred therapeutic approach. Conclusions In summary, we find that IDUS is useful for differentiating between malignant and benign biliary strictures, especially in upper and middle portions of the extrahepatic bile duct. We have also observed that in patients with strictures of the lower portion of the extrahepatic bile duct, marked echo attenuation often confounds differential diagnosis. Our experience has been that the pathologic examination of PTCS biopsy specimens is highly reliable. We have also observed that NBI facilitates detailed assessment of the changes in the biliary mucosa and is particularly useful in identifying benign and malignant disease. Finally, it is our experience that percutaneous dilation of benign strictures through the PTBD sinus tract is a safe approach to management. References 1. Fulcher AS, Turner MA, Capps G, et al. Half-Fourier RARE MRCP: experience in 300 subjects. Radiology 1998;207:21 32. 2. Inui K, Yoshino J, Okushima K, et al. Intraductal EUS. Gastrointest Endosc 2002;4:S58 S62. 3. Inui K, Miyoshi H. Cholangiocarcinoma and intraductal sonography. Gastrointest Endosc Clin N Am 2005;15:143 155. 4. Inui K, Nakazawa S, Yoshino J, et al. Biliary endoscopic findings and histological findings in bile duct carcinoma [in Japanese with English abstract]. Stomach and Intestine. 1994;29:771 775. 5. Inui K, Yoshino J, Miyoshi H, et al. Specific diagnosis of biliary strictures by quantitative assessment using a cholangioscopically derived hemoglobin index. Dig Endosc 2005;17:318 321. 6. Fujita N, Noda Y, Yokohata K, et al. Newly developed ultrasonic probe with ropeway system for transpapillary intraductal ultrasonography of the bilio-pancreatic ductal system. Dig Endosc 2000;12:250 254. 7. Linghu EQ, Cheng LF, Wang XD, et al. Intraductal ultrasonography and endoscopic retrograde cholangiography in diagnosis of extrahepatic bile duct stones: a comparative study. Hepatobiliary Pancreat Dis Int 2004;3:129 132. 8. Menzel J, Poremba C, Dietl KH, et al. Preoperative diagnosis of bile duct strictures comparison of intraductal ultrasonography with conventional endosonography. Scand J Gastroenterol 2000; 35:77 82. 9. Fujita N, Noda Y, Kobayashi G, et al. Staging of bile duct carcinoma by EUS and IDUS. Endoscopy 1998;30:A132 A134. 10. Tamada K, Inui K, Menzel J. Intraductal ultrasonography of the bile duct system. Endoscopy 2001;33:878 885. 11. Farrell RJ, Agarwal B, Brandwein SL, et al. Intraductal US is a useful adjunct to ERCP for distinguishing malignant from benign biliary strictures. Gastrointest Endosc 2002;56:681 687. 12. Stavropoulos S, Larghi A, Verna E, et al. Intraductal ultrasound for the evaluation of patients with biliary strictures and no abdominal mass on computed tomography. Endoscopy 2005;37: 715 721. 13. Kanemaki N, Nakazawa S, Inui K, et al. Three-dimensional intraductal ultrasonography: preliminary results of a new technique for the diagnosis of diseases of the pancreatobiliary system. Endoscopy 1997;29:726 731. 14. Nimura Y. Staging of biliary carcinoma: cholangiography and cholangioscopy. Endoscopy 1993;25:76 80. 15. Nimura Y. Staging cholangiocarcinoma by cholangioscopy. HPB (Oxford) 2008;10:113 115. Reprint requests Address requests for reprints to: Kazuo Inui, MD, PhD, AGAF, Department of Internal Medicine, Second Teaching Hospital, Fujita Health University School of Medicine, 3-6-10, Otobashi, Nakagawa-ku, Nagoya 454-8509, Japan. e-mail: kinui@fujita-hu.ac.jp; fax: (81) 52-323-9886. Conflict of interest The authors disclose no conflicts.