Comparison of brush and basket cytology in differential diagnosis of bile duct stricture at endoscopic retrograde cholangiopancreatography

Ki Bae Bang, Hong Joo Kim, Jung Ho Park, Dong Il Park, Yong Kyun Cho, Chong Il Sohn, Woo Kyu Jeon and Byung Ik Kim

Seoul, Korea

Comparison of brush and basket cytology in differential diagnosis of bile duct stricture at endoscopic retrograde cholangiopancreatography

Ki Bae Bang, Hong Joo Kim, Jung Ho Park, Dong Il Park, Yong Kyun Cho, Chong Il Sohn, Woo Kyu Jeon and Byung Ik Kim

Seoul, Korea

BACKGROUND:A previous report has identified a significantly higher sensitivity of cancer detection for dedicated grasping basket than brushing at endoscopic retrograde cholangiopancreatography (ERCP). This study aimed to compare the diagnostic accuracy of Geenen brush and Dormia basket cytology in the differential diagnosis of bile duct stricture.

METHOD:The current study enrolled one hundred and fourteen patients who underwent ERCP with both Geenen brush and Dormia basket cytology for the differential diagnosis of bile duct stricture at our institution between January 2008 and December 2012.

RESULTS:We adopted sequential performances of cytologic samplings by using initial Geenen brush and subsequent Dormia basket cytology in 59 patients and initial Dormia basket and subsequent Geenen brush cytology in 55 patients. Presampling balloon dilatations and biliary stentings for the stricture were performed in 17 (14.9%) and 107 patients (93.9%), respectively. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of Geenen brush cytology for the diagnosis of malignant bile duct stricture were 75.0%, 100.0%, 100.0%, 66.7% and 83.3%, respectively, and those of Dormia basket cytology were 64.5%, 100.0%, 100.0%, 58.5% and 76.3%, respectively (P=0.347 and 0.827 for sensitivity and accuracy, respectively). The good and excellent cellular yields (≥ grade 2) were obtained by Geenen brush and Dormia basket cytology in 88 (77.2%) and 79 (69.3%) patients, respectively.

CONCLUSION:The sensitivity, specificity and accuracy of biliary sampling with a Dormia basket are comparable to those with conventional Geenen brush cytology in the detection of malignant bile duct stricture.

(Hepatobiliary Pancreat Dis Int 2014;13:622-627)

bile duct stricture;

cholangiocarcinoma;

brush cytology;

basket cytology

Introduction

Cholangiocarcinoma often presents as a stricture. The accurate distinction between benign and malignant strictures of the biliary tract frequently presents a dilemma to both clinicians and pathologists for a number of reasons. Benign and malignant strictures can be indistinguishable by imaging[1]and difficult for gastroenterologists to obtain adequate intraductal forceps biopsy at the time of endoscopic retrograde cholangiopancreatography (ERCP) because of the narrow caliber of the bile duct, laterally placed lesions, and the scirrhous nature of many biliary tumors.[2-4]

Currently, available methods of biliary sampling during endoscopic drainage of suspected malignant biliary strictures have proved to be unsatisfactory.[5,6]Because no other single method has been superior, brush cytology remains the standard modality of biliary sampling at ERCP.[7]However, false negative results with brush cytology are observed in 43%-91% of cases.[3,5,8-15]Conservative management with biliary drainage is used in many patients with malignant biliary tumors, and most of them are not candidates for curative resection due to locally advanced or metastatic disease. In such patients, a concept of the process of further management can be guided by an accurate tissue diagnosis.

A previous report[16]identified a significantly higher sensitivity for cancer detection for dedicated grasping basket than brushing. However, it requires the use of dedicated grasping basket of the Bioscan catheter (T.E.I.M.E., Buenos Aires, Argentina), which is not readily available in usual clinical circumstances. Additionally, a new technique, called intraductal aspiration, was reported by Curcio et al[17]and showed a significantly higher sensitivity and superior cellular adequacy than brushing.

So far, there have been no reports comparing the diagnostic accuracy between conventional Dormia basket (which is popular and readily available) and brush cytology for the differential diagnosis of bile duct stricture during ERCP. We analyzed the prospectively collected data in our study cohort to compare the accuracy for cancer detection in biliary sampling at ERCP using a conventional Dormia basket or brushing.

Methods

Patients

One hundred and fourteen patients who underwent ERCP with both brush and Dormia basket cytology for the differential diagnosis of bile duct stricture(s) at our institution between January 2008 and December 2012 were enrolled in the current study. Patients were considered eligible if they had a suspected malignant or benign obstruction of the extra- and/or intrahepatic bile duct without prior histopathological confirmation. Exclusion criteria included an age under 18 years and coagulopathy not corrected by the administration of vitamin K or fresh frozen plasma. The final diagnosis was made based on the results of pathology of surgery and/or endobiliary forceps biopsy, endoscopic ultrasound guided fine needle aspiration cytology and biopsy (EUSFNAB) and clinical follow-up [any deterioration shown in follow-up computed tomography (CT) or magnetic resonance imaging (MRI), and clinical symptoms and signs obtained during over 12 months of follow-up]. All procedures, including obtaining written informed consent from the patient or a responsible relative, were conducted in accordance with the recommendations of the Ethics Committee of the Kangbuk Samsung Hospital (KBC11057, approved at 2011-05-02).

Interventions

Conventional ERCP and tissue acquisition procedures were performed by a single experienced endoscopist. When the biliary strictures were tight enough and could not be passed by Geenen brush or Dormia basket, the biliary strictures were dilated using the 10 mm CRETM wire-guided balloon dilator (Boston Scientific Corp., Natick, MA, USA). A Dormia basket was inserted into the bile duct above the stricture and was fully deployed at this level. Then it was withdrawn through the stricture in the open position, pulled into the tip of the catheter once it was located immediately below the stricture (this procedure was repeated at least 10 times). The unit basket/catheter was finally removed. Biliary brushing was performed with a standard 6F Geenen spring tip sheathed cytology brush (Wilson-Cook Medical Inc., Winston-Salem, NC, USA). The brush/sheath was passed as a unit through the stricture, withdrawn immediately below the stricture, and the brush was moved back and forth within the stricture segment at least 10 times. Sequential performances of brush and Dormia basket cytology were adopted in the current study. The sequence of Geenen brushing followed by Dormia basket cytology was chosen in 59 patients and the sequence of Dormia basket and then Geenen brushing cytology were chosen in 55 patients. The order of the sequence of Geenen brushing and Dormia basket cytology was determined by alternative allotment, such as oddnumbered patient who was allocated to the sequence of Geenen brushing followed by Dormia basket cytology. Even-numbered patient was allocated to the sequence of Dormia basket and then Geenen brushing cytology. Because of the retrospective design of our study, some overlapping of above-mentioned sequences is present. Endobiliary forceps biopsy was performed after the above-mentioned sequence of Geenen brushing and Dormia basket cytology if the biliary stricture could be accessible by conventional biopsy forceps. No introducer for endobiliary forceps biopsy was used in the current study.

Cytopathological examination

The cellular materials adherent to the brush and basket were directly transferred to a glass slide in the endoscopy room. An on-site cytopathologist was not available in our institution, so rapid on-site evaluation for cytologic specimens could not be done. The cytologic specimens were subsequently stored in 50% ethanol and processed to our institute of pathology. Smears were than stained using Papanicolaou's method. Specimens were reported as (a) negative for malignancy, (b) atypical considered reactive, (c) highly atypical suspicious for malignancy, (d) malignant, or (e) insufficient for diagnosis (if fewer than five clusters with ≥10 wellpreserved cells per cluster on a minimum of two slides were present). Specimens reported as negative for malignancy, atypical considered reactive, and insufficient for diagnosis were combined as negative for malignancy,and highly atypical suspicious of malignancy and malignant categories were combined as positive for malignancy.[3,8-11,13,15,16]This strategy was determined upon agreement of all authors before the initiation of this study to increase sensitivity for malignancy detection, while maintaining specificity (a cytopathologic diagnosis of "highly atypical suspicious for malignancy" together with imaging studies compatible with a malignancy are considered as sufficient evidences of malignancy by oncologists and a radiation oncologist at our institution to administer chemo- or radiotherapy). A determination of the cellular yield for each slide of brushing or Dormia basket cytology was made according to the established criteria.[18]Two pathologists in our institution blinded to the clinical data independently evaluated the cytologic specimens obtained by Geenen brushing and Dormia basket. If there were discrepancies between them, the results were read again by both pathologists together.

Statistical analysis

Statistical analysis was made using the PASW statistics package 18 (IBM, Armonk, NY, USA). Continuous variables were expressed as mean and standard deviation (SD) and were compared using independent samplesttest or the Mann-WhitneyUtest in the setting of nonparametric test. Categorical variables were compared using the the Chi-square test or Fisher's exact test. All tests were two-sided, andPvalues less than 0.05 were considered statistically significant.

Table 1.Demographics and clinical characteristics of the study participants (n=114)

Results

The epidemiologic and clinical characteristics of enrolled patients are listed in Table 1. From January 2008 to December 2012, 114 consecutive patients with biliary strictures underwent cytologic samplings by brush and Dormia basket, yielding a total of 231 cytologic specimens. Sequential performances of cytologic samplings were adopted in the current study. Specimens were taken from the biliary stricture site of 59 patients by initial Geenen brush and subsequent Dormia basket, and the protocol for initial Dormia basket and subsequent Geenen brush cytology was adopted in 55 patients. During the study period, there were no cases of any technical failure or failure to pass a stricture from two cytologic sampling methods. Endobiliary forcep biopsy was possible in 75 patients (65.8%). The final diagnoses were confirmed by pathological specimens obtained at surgery (n=35, 30.7%), forcep biopsy during ERCP (n=24, 21.1%), EUS-FNAB (n=9, 7.9%) and clinical follow-up (any deterioration in follow-up CT or MRI scan, and clinical symptoms and signs) for over 12 months (n=46, 40.4%). A relatively large number of patients with benign stricture was included in the current study (38/114, 33.3%). Bile duct drainage was completed after samplings in 107 (93.9%) patients by inserting a stent.

The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy for the detection of malignant bile duct stricture with the first sample (78.6%, 100.0%, 100.0%, 70.4% and 86.0%) collected in each patient was nearly identical to those for the second sample (71.1%, 100.0%, 100.0%, 63.3% and 80.7%). The sensitivity, specificity, PPV, NPV and accuracy of Geenen brush and Dormia basket cytology to diagnose the malignant bile duct stricture were 75.0%, 100.0%, 100.0%, 66.7%, 83.3% and 64.5%, 100.0%, 100.0%, 58.5%, 76.3%, respectively (P=0.347 and 0.827 for sensitivity and accuracy, respectively). The sensitivity, specificity, PPV, NPV and accuracy of endobiliary forceps biopsy for the diagnosis of malignant bile duct stricture were 75.0%, 100.0%, 100.0%, 57.5% and 81.3%, and were comparable to those of brush and Dormia basket, respectively (Table 2). The good and excellent cellular yields (≥grade 2) were obtained in 88 (77.2%) and 79 (69.3%) patients who underwent brush and Dormia basket cytology, respectively (Table 3).

Balloon dilatation was performed in 17 (14.9%)patients with final diagnosis of malignant (n=14) and benign bile duct stricture (n=3) before brush and basket cytology. Presampling stricture dilatation was associated with a higher sensitivity for cancer detection; however, it was not statistically significant [12/14 (85.7%) vs 45/62 (72.6%) for brushing, P=0.134; 9/14 (64.3%) vs 40/62 (64.5%) for basket cytology, P=1.000].

During the sampling procedures, no extrabiliary leakage of contrast medium or significant intraductal bleeding was detected. Complications within one month of biliary sampling were detected in 6 patients (30-day complication rate 5.3%). They consisted of 4 patients with mild acute pancreatitis associated with abdominal pain and more than 3-fold elevation of serum amylase/ lipase that resolved after conservative medical treatment and 2 patients with post-endoscopic sphincterotomy bleeding. One episode of post-ERCP bleeding was successfully controlled by local injection of 3% salineepinephrine (10:1 volume ratio mixed solution). However, massive post-ERCP bleeding in one patient was not successfully controlled by local injection of 3% salineepinephrine, and subsequent angiographic embolization of the gastroduodenal artery was required to control the bleeding.

Table 2. Cytopathological diagnoses made with 71 pairs of samples (one pair per patient) (n, %)

Table 3. Comparisons of grading for cellular yield between Dormia basket and brush cytologic sampling (n, %)

Discussion

In the current study, Dormia basket cytology sampling provided a comparable sensitivity, specificity, and accuracy for cancer detection to Geenen brush cytology in both sequences of cytologic samplings. Biliary brush cytology, first described by Osnes et al[19]in 1975, is the most commonly used tissue sampling technique when a biliary stricture suspicious for malignancy is encountered at ERCP. Although it has a specificity close to 100%, brush cytology has a modest sensitivity for detection of cancer, ranging from 18% to 60% in most published series.[1-10]It has been suggested that this limited sensitivity is at least partially attributed to failure to obtain an adequate cellular yield. This may be due to submucosal tumor spread or an extrinsic location of the tumor with compression of the biliary tree.[20,21]In the current study, the sensitivity of Geenen brush cytology for the detection of malignant bile duct stricture was 75.0% (57 positive for malignancy out of 76 total patients with malignant bile duct strictures). In addition, the equivalent high sensitivity of Dormia basket cytology for the detection of malignant bile duct stricture compared to Geenen brush cytology was also noted in the current study. The sensitivities of the present study are somewhat superior to those of previous studies. The reasons of higher sensitivity in our study are speculated to be as follows: i) the number (n=61) of enrolled patients diagnosed finally with malignant bile duct stricture due to cholangiocarcinoma and gallbladder cancer was larger than that (n=15) of those with bile duct stricture due to the extrinsic compression such as pancreatic adenocarcinoma, metastasis and intraductal papillary mucinous carcinoma, and ii) relatively good and excellent cellular yields (≥grade 2) were obtained by both brush and Dormia basket cytology (77.2% and 69.3%, respectively).

In the current study, we adopted sequential performances of cytologic samplings in which initial Geenen brush cytology and subsequent Dormia basket cytology were taken in 59 patients and initial Dormia basket cytology and subsequent Geenen brush cytology in 55 patients. Original anticipation which was inferred from the previous study[16]was that the sensitivity for the detection of malignant bile ductstricture by initially taken cytological specimens might be superior to the subsequently taken cytological specimens. However, the sensitivity, specificity, PPV, NPV and accuracy for the detection of malignant bile duct stricture in the first sample collected in each patient were not significantly different from those in the second sample. Although the sensitivity, specificity, PPV, NPV and accuracy were similar between the first and second cytological examinations, positive results for malignancy (cytological grades 3 and 4) could be obtained in the second cytologic examination after the negative results for malignancy in the first cytologic examination in 3 (2.6%) of our enrolled patients. We can cautiously suppose that additional yields can be obtained by the second cytological sampling in patients with indeterminate bile duct stricture.

Dumonceau et al[16]reported that biliary cytology sampling using the dedicated basket grasping and presampling stricture dilatation provided a significantly higher sensitivity for cancer detection than brushing. They used the grasping basket of a Bioscan catheter, which is made of stainless steel and about 20 mm in diameter. Theoretically, this dedicated grasping basket is better than a brush in terms of the diameter and stiffness, which allows a better grasp of deep tissue exposed by dilation. However, this dedicated grasping basket which is not readily available in the clinical setting is more difficult to handle because the diameter of the basket is larger than that of a conventional Dormia basket. To our knowledge, there have been no reports on the clinical use of a conventional Dormia basket, which is readily available in the clinical setting, to grasp or abrase the bile duct stricture lesion and to compare the diagnostic efficacy of Dormia basket with that of conventional brush cytology. In the current study, we found that the sensitivity, specificity and diagnostic accuracy of Dormia basket were not inferior to the conventional brush cytology. In addition, the cellular yield of Dormia basket cytology, which may influence the diagnostic efficacy of the technique, was also comparable to that of the conventional brush cytology.

In our study, there were no significant differences in the technical difficulty of brush and Dormia basket cytology. They are both easily passed through the stricture lesion of the bile duct, and could also be easily moved back and forth/deployed and withdrawn in most of our enrolled patients. Because of the straight and stiff end of the Geenen brush, however, we anticipate more technical difficulties in passing the stricture lesion and moving back and forth with Geenen brush than Dormia basket, especially in the curved stricture lesion as in mid to distal common bile duct and hilar portions, without the aid of a guidewire technique. Actually, we considered that the straight tip of Geenen brush may hinder the passage of some curved regions of the extrahepatic bile duct. In these cases, we adopted the wire-guided insertion of Geenen brush and performed cytologic examinations. Using Dormia basket, we easily passed the stricture lesions and also easily performed deploy/withdraw actions due to the curved and flexible tip and more efficient delivery of the pushing force than Geenen brush. Further gathering of clinical data for these technical aspects will be warranted.

In our study, presampling balloon dilatation was performed in 17 (14.9%) patients with a final diagnosis of malignant (n=14) and benign bile duct stricture (n=3). Presampling stricture dilatation was associated with a higher sensitivity for cancer detection; however, it was not statistically significant. Likewise, a previous study by de Bellis et al[11]also showed that stricture dilatation did not improve the sensitivity of brush cytology. Conversely, Farrell et al[10]reported that the combination of stricture dilatation, and subsequent biliary brushing significantly increased both the sensitivity and specificity of cytology with positive brushings in all patients with pancreatic or gallbladder carcinoma. Further clinical trials will be needed in this area.

Endobiliary forceps biopsy is not always possible in patients with bile duct stricture. In our study, endobiliary forceps biopsy was possible in only 75 (65.8%) of our enrolled patients. The sensitivity, specificity, PPV, NPV and accuracy of endobiliary forceps biopsy for the diagnosis of malignant bile duct stricture were comparable to those of brush and Dormia basket cytology, respectively (Table 2). A previous report[22]showed a similar feasibility rate of endobiliary forcep biopsy (530/766, 69.2%) in bile duct stricture. Additionally, a nearly identical sensitivity (74.4%) of endobiliary forceps biopsy for the detection of malignant bile duct stricture was reported in the same study.

In conclusion, biliary sampling with a Dormia basket provided sensitivity, specificity and accuracy for the detection of malignant bile duct stricture comparable to conventional Geenen brush cytology.

Contributors:BKB and KHJ proposed the study, performed research, collected and analyzed the data and wrote the first draft. All authors contributed to the design and interpretation of the study and to further drafts. KHJ is the guarantor.

Funding:This study was financially supported by a grant from JEIL Pharmaceuticals (Seoul, Korea).

Ethical approval:All procedures, including obtaining written informed consent from the patient or a responsible relative, were conducted in accordance with the recommendations of the Ethics Committee of the Kangbuk Samsung Hospital (KBC11057).

Competing interest:The author or one or more of the authors have received for personal or professional use from a commerical party related directly or indirectly to the subject of this article.

1 Lindberg B, Arnelo U, Bergquist A, Th?rne A, Hjerpe A, Granqvist S, et al. Diagnosis of biliary strictures in conjunction with endoscopic retrograde cholangiopancrea ticography, with special reference to patients with primary sclerosing cholangitis. Endoscopy 2002;34:909-916.

2 Ponchon T, Gagnon P, Berger F, Labadie M, Liaras A, Chavaillon A, et al. Value of endobiliary brush cytology and biopsies for the diagnosis of malignant bile duct stenosis: results of a prospective study. Gastrointest Endosc 1995;42: 565-572.

3 Jailwala J, Fogel EL, Sherman S, Gottlieb K, Flueckiger J, Bucksot LG, et al. Triple-tissue sampling at ERCP in malignant biliary obstruction. Gastrointest Endosc 2000;51:383-390.

4 Pugliese V, Conio M, Nicolò G, Saccomanno S, Gatteschi B. Endoscopic retrograde forceps biopsy and brush cytology of biliary strictures: a prospective study. Gastrointest Endosc 1995;42:520-526.

5 Moreno Luna LE, Kipp B, Halling KC, Sebo TJ, Kremers WK, Roberts LR, et al. Advanced cytologic techniques for the detection of malignant pancreatobiliary strictures. Gastroenterology 2006;131:1064-1072.

6 Kozarek RA. Analog diagnosis in a digital world: the quest to improve endoscopic tissue acquisition and diagnostic sensitivity in malignant biliary stricture. Clin Gastroenterol Hepatol 2004;2:207-208.

7 Eisen GM, Dominitz JA, Faigel DO, Goldstein JL, Kalloo AN, Petersen BT, et al. An annotated algorithmic approach to malignant biliary obstruction. Gastrointest Endosc 2001;53: 849-852.

8 Mansfield JC, Griffin SM, Wadehra V, Matthewson K. A prospective evaluation of cytology from biliary strictures. Gut 1997;40:671-677.

9 Glasbrenner B, Ardan M, Boeck W, Preclik G, M?ller P, Adler G. Prospective evaluation of brush cytology of biliary strictures during endoscopic retrograde cholangiopancreatog raphy. Endoscopy 1999;31:712-717.

10 Farrell RJ, Jain AK, Brandwein SL, Wang H, Chuttani R, Pleskow DK. The combination of stricture dilation, endoscopic needle aspiration, and biliary brushings significantly improves diagnostic yield from malignant bile duct strictures. Gastrointest Endosc 2001;54:587-594.

11 de Bellis M, Fogel EL, Sherman S, Watkins JL, Chappo J, Younger C, et al. Influence of stricture dilation and repeat brushing on the cancer detection rate of brush cytology in the evaluation of malignant biliary obstruction. Gastrointest Endosc 2003;58:176-182.

12 R?sch T, Hofrichter K, Frimberger E, Meining A, Born P, Weigert N, et al. ERCP or EUS for tissue diagnosis of biliary strictures? A prospective comparative study. Gastrointest Endosc 2004;60:390-396.

13 Fogel EL, de Bellis M, McHenry L, Watkins JL, Chappo J, Cramer H, et al. Effectiveness of a new long cytology brush in the evaluation of malignant biliary obstruction: a prospective study. Gastrointest Endosc 2006;63:71-77.

14 Harewood GC, Baron TH, Stadheim LM, Kipp BR, Sebo TJ, Salomao DR. Prospective, blinded assessment of factors influencing the accuracy of biliary cytology interpretation. Am J Gastroenterol 2004;99:1464-1469.

15 Howell DA, Parsons WG, Jones MA, Bosco JJ, Hanson BL. Complete tissue sampling of biliary strictures at ERCP using a new device. Gastrointest Endosc 1996;43:498-502.

16 Dumonceau JM, Macias Gomez C, Casco C, Genevay M, Marcolongo M, Bongiovanni M, et al. Grasp or brush for biliary sampling at endoscopic retrograde cholangiography? A blinded randomized controlled trial. Am J Gastroenterol 2008;103:333-340.

17 Curcio G, Traina M, Mocciaro F, Liotta R, Gentile R, Tarantino I, et al. Intraductal aspiration: a promising new tissue-sampling technique for the diagnosis of suspected malignant biliary strictures. Gastrointest Endosc 2012;75:798-804.

18 Camp R, Rutkowski MA, Atkison K, Niedzwick L, Vakil N. A prospective, randomized, blinded trial of cytological yield with disposable cytology brushes in upper gastrointestinal tract lesions. Am J Gastroenterol 1992;87:1439-1442.

19 Osnes M, Serck-Hanssen A, Myren J. Endoscopic retrograde brush cytology (ERBC) of the biliary and pancreatic ducts. Scand J Gastroenterol 1975;10:829-831.

20 Layfield LJ, Wax TD, Lee JG, Cotton PB. Accuracy and morphologic aspects of pancreatic and biliary duct brushings. Acta Cytol 1995;39:11-18.

21 Logrono R, Kurtycz DF, Molina CP, Trivedi VA, Wong JY, Block KP. Analysis of false-negative diagnoses on endoscopic brush cytology of biliary and pancreatic duct strictures: the experience at 2 university hospitals. Arch Pathol Lab Med 2000;124:387-392.

22 Kim JY, Choi JH, Kim JH, Kim CL, Bae SH, Choi YK, et al. Clinical usefulness of bile cytology obtained from biliary drainage tube for diagnosing cholangiocarcinoma. Korean J Gastroenterol 2014;63:107-113.

Received January 17, 2014

Accepted after revision May 30, 2014

Author Affiliations: Department of Internal Medicine, Sungkyunkwan University Kangbuk Samsung Hospital, 108, Pyung-Dong, Jongro-Ku, Seoul 110-746, Korea (Bang KB, Kim HJ, Park JH, Park DI, Cho YK, Sohn CI, Jeon WK and Kim BI)

Hong Joo Kim, MD, Department of Internal Medicine, Sungkyunkwan University Kangbuk Samsung Hospital, 108, Pyung-Dong, Jongro-Ku, Seoul 110-746, Korea (Tel: +82-2-2001-8556; Fax: +82-2-2001-2049; Email: hongjoo3.kim@samsung.com)

? 2014, Hepatobiliary Pancreat Dis Int. All rights reserved.

10.1016/S1499-3872(14)60311-8

Published online October 27, 2014.