Intraductal aspiration: a promising new tissue-sampling technique for the diagnosis of suspected malignant biliary strictures

ORIGINAL ARTICLE: Clinical Endoscopy

Intraductal aspiration: a promising new tissue-sampling technique for the diagnosis of suspected malignant biliary strictures Gabriele Curcio, MD,1 Mario Traina, MD,1 Filippo Mocciaro, MD,1 Rosa Liotta, MD,2 Raffaella Gentile, MD,2 Ilaria Tarantino, MD,1 Luca Barresi, MD,1 Antonino Granata, MD,1 Fabio Tuzzolino, MS,3,5 Bruno Gridelli, MD4 Palermo, Italy

Background: Brushing is the most commonly used technique for biliary sampling at ERCP, despite its limited sensitivity. Objective: To evaluate intraductal aspiration (IDA) as a new combined endoscopic technique for cytodiagnosis, its cellular adequacy, diagnostic accuracy for cancer detection, feasibility, and safety. Design: Prospective, observational study. Setting: Single tertiary referral center. Main Outcome Measurements: IDA cellular adequacy, diagnostic accuracy for cancer detection, feasibility, and safety. Patients and Methods: From April 2009 to September 2010, 42 consecutive patients with suspected malignant biliary stricture underwent ERCP, with tissue sampling obtained with IDA. IDA included performance of standard brushing in all patients. After standard brushing, to perform IDA, we removed the brush from its catheter and used the tip of the catheter as a scraping device. The tip was scraped back and forth across the stricture at least 10 times. The catheter and a suction line were connected to a specimen trap to obtain intraductal aspiration of fluids and samplings. Results: Our cytopathologists found adequate cellular yield in 39 of the 42 IDA samples (92.8%) versus 15 of the 42 brushing samples (35.7%) (P ⬍ .001). IDA showed a significantly higher sensitivity than brushing (89% vs 78% for adequate samples and 89% vs 37% for all samples) and provided significantly superior cellular adequacy (92.8% vs 35.7%). Limitations: Observational study, small number of patients. Conclusions: IDA significantly improves brushing cellular adequacy and has high sensitivity for cancer detection. It was also safe, simple, rapid, and applicable during routine diagnostic ERCP, with no additional costs. (Gastrointest Endosc 2012;75:798-804.)

Biliary strictures are a common occurrence in clinical practice. Although many of these strictures are caused by malignancies of the biliary tract, they may also have a nonmalignant etiopathogenesis, such as inflammatory

conditions, choledocholithiasis, chronic pancreatitis, surgical trauma, and ischemia. Diagnostic tissue acquisition at ERCP may obviate the need for further invasive testing, thus allowing for optimal

Abbreviations: H&E, hematoxylin and eosin; IDA, intraductal aspiration; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver-operating characteristic; ROSE, rapid on-site evaluation.

Current affiliations: Departments of Endoscopy (1), Pathology (2), Information Technology (3), and Transplantation Surgery (4), Mediterranean Institute for Transplantation and Advanced Specialized Therapies (IsMeTT), Palermo, Italy, Department of Mathematics and Statistics (5), S. Vianelli University of Palermo, Palermo, Italy.

DISCLOSURE: The authors disclosed no financial relationships relevant to this publication. Copyright © 2012 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 doi:10.1016/j.gie.2011.12.005 Received August 9, 2011. Accepted December 5, 2011.

798 GASTROINTESTINAL ENDOSCOPY Volume 75, No. 4 : 2012

Reprint requests: Gabriele Curcio, MD, Department of Gastroenterology, IsMeTT, UPMC, Via Tricomi 1, Palermo 90100, Italy. If you would like to chat with an author of this article, you may contact Dr Curcio at [email protected].

www.giejournal.org

Curcio et al

Intraductal aspiration

TABLE 1. Patient characteristics at baseline Characteristic Age ⫾ SD, y

Patients (N ⴝ 42) 61.9 ⫾ 18.6

Male sex, no. (%)

28 (66.7)

Symptomatic at presentation, no. (%)

40 (95.2)

Take-home Message ●

Intraductal aspiration significantly improves brushing cellular adequacy and has a high sensitivity for cancer detection. It is also safe, simple, rapid, and applicable during routine diagnostic ERCP.

Type of biliary stricture, no. (%) Upper third

16 (38.1)

Intrahepatic

12 (28.6)

Middle third

8 (19)

Lower third

6 (14.3)

Stricture dilation performed before sampling, no. (%)

5 (11.1)

SD, Standard deviation.

intervention without delay.1 Adequacy of specimens is essential for tissue diagnosis, although several studies showed that specimens obtained from a biliary stricture may lack adequate cellularity, resulting in a false-negative diagnosis regardless of the technique used.2 Because no other single method has been found to be superior, brushing remains the most commonly used technique for biliary sampling at ERCP, despite its limited sensitivity.3-6 The challenge in improving diagnosis of indeterminate strictures is devising new and more effective ways to target lesions and retrieve higher-quality cytology specimens. We evaluated the feasibility of intraductal aspiration (IDA) as a new combined endoscopic technique (brushing plus scraping and aspiration) for cytodiagnosis of suspected malignant biliary strictures. We evaluated the cellular adequacy of samples, diagnostic accuracy, and the safety of this new technique.

MATERIALS AND METHODS Patients From April 2009 to September 2010, after approval from our internal review board, 42 consecutive adult patients with suspected malignant biliary stricture and no previous histological confirmation underwent ERCP at our institution. Tissue samples were obtained with IDA in all patients. The IDA technique included performance of standard brushing, as specified in the following. Patient characteristics are summarized in Table 1. Twenty-eight male patients (66.7%) and 14 female patients (33.3%) with a mean ⫾ standard deviation age of 61.9 ⫾ 18.6 years were enrolled. The majority of patients (95.2%) were symptomatic (eg, jaundice, abdominal pain, fever, pruritus). Informed consent for ERCP, including acquisition of cytology specimens with IDA, was obtained in writing www.giejournal.org

from all patients. In all patients, routine diagnostic procedures included laboratory tests, abdominal US, abdominal CT scan, and cholangio-magnetic resonance imaging. With the patient under general anesthesia or deep sedation, ERCP with tissue sampling was performed, with the patient continuously monitored by electrocardiography, pulse oximetry, and automatic recording of blood pressure and pulse. All patients were closely followed for any early or delayed complications related to the procedure and until the final diagnosis. Patients with no finding of malignancy were also closely followed with clinical observation, laboratory tests, and imaging. Only patients with persistent suspicion of malignancy, despite negative cytology findings for cancer, underwent further cytopathological sampling at ERCP. Patients with no finding of malignancy during a minimum of 12 months of follow-up were considered negative for a final diagnosis of malignancy.

Intervention and technique ERCP was performed with a standard videoduodenoscope (TFJ 160-VR; Olympus, Hamburg, Germany). To perform IDA, bile duct strictures were identified under radiographic guidance and by using contrast fluid. A 0.035-inch guidewire (Jagwire; Boston Scientific, Natick, Mass) was then advanced through the stricture. Over the guidewire, we did standard brushing of the stricture by using the RX Wire-Guided Cytology Brush, 2.8 mm in diameter (Boston Scientific) (Fig. 1A). The brush was advanced from the catheter to a point proximal to the stricture, withdrawn slightly, and moved back and forth across the stricture at least 10 times. The brush was then pulled into the tip of the catheter, still located immediately below the stricture, and the brush/catheter unit was finally removed, leaving the guidewire in place. We then pushed the brush so that it protruded from the catheter and cut for smearing. After that, we withdrew the stylet from the catheter, which we then used to perform the aspiration, by using the tip as a scraping device (Fig. 1B). This modified catheter was advanced over the guidewire and moved back and forth, scraping across the stricture at least 10 times. During the IDA procedure, both the catheter and a specimen trap were connected to the central suction line (0.6 Bar) to obtain intraductal aspiration of fluids and samples (Fig. 1C). After biliary sampling, a plastic stent for drainage was placed in each patient. Volume 75, No. 4 : 2012 GASTROINTESTINAL ENDOSCOPY 799

Intraductal aspiration

Curcio et al

Figure 1. Standard brushing catheter (A) before and (B) after removing the brush. C, Intraductal aspiration catheter and suction line connected to a specimen trap.

Cytopathological examination

Study endpoints

The cellular material adhering to the brush was directly smeared, wet-fixed, and stained with rapid hematoxylin and eosin (H&E), allowing the cytopathologist to do a rapid on-site evaluation (ROSE) and assess adequacy. The brush was subsequently stored in 50% ethanol and processed in our pathology laboratory to collect the remaining material by cytospin. IDA fluid was stored in 50% ethanol. A drop of IDA fluid was directly smeared, wet-fixed, and stained with rapid H&E to perform ROSE. ROSE allowed interpretation of samples as adequate or inadequate based on the cellular yield for both sets, ensuring at least 1 interpretable specimen with brushing and/or IDA for each patient. Smears of cell pellets were done for both sets of samples, and cytospin preparations were done for both sets, when necessary, and stained with both the Papanicolaou method and H&E. When an adequate sample was available, a cell block was done, and sections were stained with H&E and, occasionally, also with special or immunohistochemical stains to typify the cells. Both brushing and IDA samples were jointly interpreted by our 2 cytopathologists, both experienced in GI cytopathology. Specimens were reported as inadequate or unsatisfactory when acellular/hypocellular, or when showing poor fixation or crush artifacts, excessive blood, necrosis, or debris that obscured cellular details. Adequate specimens were reported as normal, atypical considered reactive, highly atypical suspicious for cancer, and malignant. Specimens reported as benign or atypical considered reactive were combined as negative for cancer, and the highly atypical suspicious for cancer and malignant categories were combined as positive for cancer.7,8 A cytopathological diagnosis of highly atypical suspicious for cancer, together with imaging studies compatible with a malignancy, was considered sufficient evidence of malignancy for consequent surgical or medical treatment.

The study endpoints were to evaluate IDA for cellular adequacy of the samples, diagnostic accuracy for cancer detection, feasibility, and safety. All of the results were compared with those of standard brushing alone.

800 GASTROINTESTINAL ENDOSCOPY Volume 75, No. 4 : 2012

Statistical analysis All collected data were recorded in an electronic database. Data were analyzed by using the SPSS 15 software package (SPSS Inc, Chicago, Ill). Continuous variables were summarized as means ⫾ standard deviation. Categorical variables were summarized as frequency and percentage. Significant difference was calculated for the cellular adequacy by using a ␹2 test. Difference was considered significant at a P value of ⬍.05. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy of both techniques were calculated.

RESULTS From April 2009 to September 2010, 42 consecutive patients presenting with suspected malignant biliary stricture and no previous histological confirmation were enrolled and prospectively followed until final diagnosis. IDA was successful for all strictures, with no need for further attempts because of clogging of catheter/suction line system, possibly because of the back and forth movement of the catheter during IDA. Regarding the site of the biliary strictures, 16 (38.1%) were in the upper third, 12 (28.6%) were intrahepatic, 8 (19%) were in the middle third, and 6 (14.3%) in the lower third. Because of the severity of the stricture, balloon dilation was performed in 5 patients (11.1%) with biliary stricture before tissue sampling. Our cytopathologists found adequate cellular yield for diagnosis in 39 of the 42 IDA samples (92.8%). Seventeen of these were judged positive for cancer, with 12 maligwww.giejournal.org

Curcio et al

Intraductal aspiration

Primary sclerosing cholangitis

8 (35)

Postoperative biliary stricture

7 (30)

of brushing was 87%. On the second analysis, considering all patients, regardless of adequacy, the sensitivity of brushing alone was 37% and the specificity was 26%, with a PPV of 29% and an NPV of 33%. Overall accuracy was 31% (Table 4). Comparing the results only for patients with adequate samples, the difference in sensitivity between brushing alone and IDA was not statistically significant (P ⫽ .4753), as calculated with receiver-operating characteristic (ROC) curves (brushing ROC area 0.8889, IDA ROC area 0.9474). On second analysis, regardless of adequacy and including all patients, the difference in sensitivity between brushing and IDA was statistically significant (P ⬍ .001), as calculated with ROC curves (brushing ROC area 0.3146, IDA ROC area 0.8822).

Chronic pancreatitis

5 (22)

Safety

Infective

3 (13)

No adverse events related to IDA were recorded during the entire follow-up.

TABLE 2. Final diagnoses in 42 patients with malignant biliary stricture No. (%) Malignant, 19 patients Cholangiocarcinoma

14 (74)

Metastases*

3 (16)

Gallbladder cancer

1 (5)

Intraductal papillary mucinous tumor

1 (5)

Benign, 23 patients

*Primary tumors were colon (n ⫽ 2) and stomach (n ⫽ 1) cancer.

DISCUSSION nant (71%) and 5 highly atypical suspicious (29%). In 3 patients, the IDA sample had inadequate cellular yield, although the yield was adequate in the brushing sample. A final diagnosis of malignant stricture was made in 19 patients (45.2%). The final diagnoses in all patients are shown in Table 2. The mean follow-up for considering a patient negative for a final diagnosis of malignancy was of 16.6 ⫾ 4.2 months. In 5 patients negative for malignancy (3 with primary sclerosing cholangitis and 2 with chronic pancreatitis), ERCP with biliary sampling was repeated because there was a high suspicion of malignancy. None of these patients were found positive for malignancy at the end of follow-up. IDA showed a sensitivity for cancer detection of 89% and a specificity of 100%, with a PPV of 100% and an NPV of 91%. Overall accuracy of IDA was 95% (Table 3). We also did an intention-based diagnosis analysis, including those samples judged inadequate by the cytopathologists. Assuming that all inadequate samples can never provide a correct final diagnosis, we assigned all those samples a sensitivity, specificity, and accuracy of 0. On this second analysis, IDA showed a sensitivity for cancer detection of 89%, and a specificity of 87%, with a PPV of 85% and an NPV of 91%. Overall accuracy of IDA was 88% (Table 4). To be thorough, we calculated the same diagnostic outcome for the standard brushing alone, which remains our standard of care for tissue sampling. Our cytopathologists found adequate cellular yield for diagnosis in only 15 of the 42 brushing samples (35.7%) obtained with brushing alone. Seven of these were judged positive for cancer, with 4 malignant (57%) and 3 highly atypical suspicious (43%). The adequacy was significantly lower than for IDA (92.8%, P ⬍ .001). The sensitivity for cancer detection with brushing alone was 78% and the specificity was 100%, with a PPV of 100% and an NPV of 75%. Overall accuracy www.giejournal.org

Our results revealed that this new combined technique, which we call IDA, significantly improved cellular adequacy, showing a high sensitivity for cancer detection. Tissue collection during ERCP is widely used for distinguishing between benign and malignant strictures and for providing a definitive diagnosis that can be used for establishing therapeutic strategies. However, tissue sampling at ERCP has always been challenging. Biliary brushing is the most frequently used technique. It can be performed without technical difficulty for most biliary strictures detected at ERCP, although it has limited cytological sensitivity, which is often blamed on its low cellular yield. Although this technique has a specificity of nearly 100%, it has only modest cancer sensitivity, ranging from 18% to 60% in most published series.4-6 Tissue samples for histological investigation can be also obtained from biliary strictures by using forceps. This technique is more challenging and time-consuming than brushing and is less widely used, but it provides a sample of subepithelial stroma. The popular combination of brush cytology and forceps biopsy has a sensitivity ranging from 52.0% to 70.4%, which is slightly higher than either individual approach.9,10 A number of other sampling techniques have been evaluated, with no evidence of an ideal technique in terms of sensitivity, specificity, cost-effectiveness, and safety. Intraductal bile aspiration cytology,2 stricture dilation before brushing,11 the use of a basket after dilation,12 the use of a guidewire to scrape the stricture,13 a new technique of tissue specimen preparation (Smash protocol),14 and the use of different brush lengths and stiffness5 are reported in the literature. Other studies have evaluated directed tissue biopsy samples obtained with miniature cholangioscopic biopsy Volume 75, No. 4 : 2012 GASTROINTESTINAL ENDOSCOPY 801

Intraductal aspiration

Curcio et al

TABLE 3. Sensitivity of brushing and IDA calculated for adequate samples Cellular adequacy, no. (%)

Sensitivity, % (95% CI)

Specificity, % (95% CI)

NPV, % (95% CI)

PPV, % (95% CI)

Accuracy, %

IDA*

39 (92.8)

89 (65-98)

100 (80-100)

91 (69-98)

100 (77-100)

95

Brushing†

15 (35.7)

78 (40-96)

100 (52-100)

75 (36-96)

100 (56-100)

87

IDA, Intraductal aspiration; CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value. *True positive, n ⫽ 17; false positive, n ⫽ 0; true negative, n ⫽ 20; false negative, n ⫽ 2. †True positive, n ⫽ 7; false positive, n ⫽ 0; true negative, n ⫽ 6; false negative, n ⫽ 2.

TABLE 4. Sensitivity of brushing and IDA calculated for all samples Cellular adequacy, no. (%)

Sensitivity, % (95% CI)

Specificity, % (95% CI)

NPV, % (95% CI)

PPV, % (95% CI)

Accuracy, %

IDA*

39 (92.8)

89 (65-98)

87 (65-96)

91 (69-98)

85 (61-96)

88

Brushing†

15 (35.7)

37 (17-61)

26 (11-49)

33 (14-59)

29 (13-51)

31

IDA, Intraductal aspiration; CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value. *True positive, n ⫽ 17; false positive, n ⫽ 3; true negative, n ⫽ 20; false negative, n ⫽ 2. †True positive, n ⫽ 7; false positive, n ⫽ 17; true negative, n ⫽ 6; false negative, n ⫽ 12.

forceps15 or with a single-operator peroral cholangiopancreatoscopy system known as Spyglass.16 Advanced cytological techniques for detecting malignancy in pancreatobiliary brushing specimens, including fluorescence in situ hybridization and digital image analysis, have also been evaluated and have shown a higher sensitivity compared with routine cytology.17 Recently, p53 immunocytochemical staining as an adjunct in conventional cytology has appeared useful in improving the sensitivity of brush cytology in malignant strictures.18 Data from several studies showed that combining techniques in the same ERCP session to obtain tissue samples from biliary strictures enhances cancer detection. However, multimodal sampling is more time-consuming and more difficult technically compared with the use of a single technique.6 Our aim was to develop a new combined sampling technique to improve cellular yield and cancer detection, one that is rapid and easy to perform and incurs no added costs. In our 42 patients, IDA provided a significantly higher sensitivity than brushing (89% vs 78% for adequate samples and 89% vs 37% for all samples), providing a significantly superior cellular adequacy (92.8% vs 35.7%) (Fig. 2). Cellular adequacy was evaluated on site for brushing and IDA. In only 3 patients (7.2%) did ROSE find IDA samples inadequate, although for those 3 patients, brushing samples proved adequate. We did not repeat sampling in these cases because the brushing sample was adequate and because we wanted to avoid the risk of complications. The remaining 39 IDA samples (92.8%) were adequate at ROSE. Although the sensitivity of standard brushing was acceptable, the cellular adequacy of the samples was unsat802 GASTROINTESTINAL ENDOSCOPY Volume 75, No. 4 : 2012

isfactory, with a limited overall cancer detection rate, attributable to the low number of interpretable specimens, evidenced by a sensitivity of 37% when calculated with an intention-based diagnosis analysis. In our opinion, it is of the utmost clinical importance to have the greatest possible number of interpretable samples, with the highest possible sensitivity, as we found with IDA. A number of factors that influence tissue sampling adequacy have been identified, including proximally located strictures, tight strictures that might not be fully traversed by the guidewire, sharp angulation in the distal duct that prevents the passage of biopsy forceps or brush, the presence of firm desmoplastic tumors, and the presence of intraductal debris and blood that might contaminate the specimen, particularly the brushings for cytology.19 A tissue sampling technique for use at ERCP should ensure the retrieval of good-quality cytology specimens and, ideally, have a high sensitivity for cancer detection, with absolute specificity. To be widely used, the technique should also be simple, safe, rapid, and relatively inexpensive.2 The development of a high-yield ERCP intraprocedural tissue-sampling technique could also help avoid a delay in tissue diagnosis, obviate the need for additional procedures, permit appropriate stent selection, and reduce health care costs.14 It could be hypothesized that the low cellular yield and the unsatisfactory sensitivity of standard brushing are related to the brushing itself, which may dislodge an adequate number of cells, but fail to collect them. IDA is an attempt to enhance standard brushing diagnostic power by aspirating fluids and samples previously dislodged by brushing. Furthermore, aspiration and scraping through www.giejournal.org

Curcio et al

Intraductal aspiration

Figure 2. A, Intraductal aspiration (IDA) sample composed of diagnostic fragments of malignant epithelium, as well as non-neoplastic ductal type epithelium, and normal GI tract contamination (cell block section, H&E stain, low power). B, Detail of A. Atypical back-to-back glandular formations representing moderately differentiated components of adenocarcinoma (high power). C, IDA sample composed of several atypical epithelial tissue fragments with glandular differentiation and single atypical cells in a mucinous background (cell block section, H$E stain, low power). D, Detail of C. Neoplastic cells with enlarged and pleomorphic nuclei. A few cells have a columnar shape; others have lost their polarity and show disorganization (high power).

the stricture probably allowed for more adequate samples also by exposing subepithelial malignant tissue. In addition, the IDA procedure appeared to be safe, with no evidence of adverse events. It was also simple, rapid, and applicable during routine diagnostic ERCP, by using the same brushing catheter, with no additional costs. To ensure the standard of care for all patients, standard brushing was performed before intraductal aspiration. As a result, this study does not provide sufficient information for establishing whether IDA, done by scraping the stricture and aspirating fluids without previous brushing, would have achieved the same important results. In summary, our results were very promising, showing that IDA significantly improves brushing cellular adequacy and has a high sensitivity for cancer detection. Randomized, controlled trials are needed to evaluate the potential advantages of this new technique, comparing IDA (with www.giejournal.org

and without brushing before scraping and aspiration) with standard brushing, which remains the actual standard of care for diagnostic tissue acquisition.

REFERENCES 1. NIH state-of-the-science statement on endoscopic retrograde cholangiopancreatography (ERCP) for diagnosis and therapy. NIH Consens State Sci Statements 2002;19:1-26. 2. De Bellis M, Sherman S, Fogel EL, et al. Tissue sampling at ERCP in suspected malignant biliary strictures (part 1). Gastrointest Endosc 2002; 56:552-61. 3. Eisen GM, Dominitz JA, Faigel DO, et al. An annotated algorithmic approach to malignant biliary obstruction. Gastrointest Endosc 2001;53: 849-52. 4. Nguyen K, Sing JT Jr. Review of endoscopic techniques in the diagnosis and management of cholangiocarcinoma. World J Gastroenterol 2008; 14:2995-9.

Volume 75, No. 4 : 2012 GASTROINTESTINAL ENDOSCOPY 803

Intraductal aspiration

Curcio et al

5. Fogel EL, deBellis M, McHenry L, et al. Effectiveness of a new long cytology brush in the evaluation of malignant biliary obstruction: a prospective study. Gastrointest Endosc 2006;63:71-7. 6. De Bellis M, Sherman S, Fogel EL, et al. Tissue sampling at ERCP in suspected malignant biliary strictures (part 2). Gastrointest Endosc 2002; 56:720-30. 7. Guidelines of the Papanicolaou Society of Cytopathology for fineneedle aspiration procedure and reporting. The Papanicolaou Society of Cytopathology Task Force on Standards of Practice. Diagn Cytopathol 1997;17:239-47. 8. Henke AC, Jensen CS, Cohen MB. Cytologic diagnosis of adenocarcinoma in biliary and pancreatic duct brushings. Adv Anat Pathol 2002;9: 301-8. 9. Weber A, von Weyhern C, Fend F, et al. Endoscopic transpapillary brush cytology and forceps biopsy in patients with hilar cholangiocarcinoma. World J Gastroenterol 2008;14:1097-101. 10. Farrell RJ, Jain AK, Brandwein SL, et al. The combination of stricture dilation, endoscopic needle aspiration, and biliary brushings significantly improves diagnostic yield from malignant bile duct strictures. Gastrointest Endosc 2001;54:587-94. 11. De Bellis M, Fogel EL, Sherman S, 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; 58:176-82.

12. Dumonceau JM, Macias Gomez C, Casco C, et al. Grasp or brush for biliary sampling at endoscopic retrograde cholangiography? A blinded randomized controlled trial. Am J Gastroenterol 2008;103:333-40. 13. Uehara H, Tatsumi K, Masuda E, et al. Scraping cytology with a guidewire for pancreatic-ductal strictures. Gastrointest Endosc 2009;70:52-9. 14. Wright ER, Bakis G, Srinivasan R, et al. Intraprocedural tissue diagnosis during ERCP employing a new cytology preparation of forceps biopsy (smash protocol). Am J Gastroenterol 2011;106:294-9. 15. Shah RJ, Langer DA, Antillon MR, et al. Cholangioscopy and cholangioscopic forceps biopsy in patients with indeterminate pancreaticobiliary pathology. Clin Gastroenterol Hepatol 2006;4:219-25. 16. Chen YK, Pleskow DK. SpyGlass single-operator peroral cholangiopancreatoscopy system for the diagnosis and therapy of bile-duct disorders: a clinical feasibility study (with video). Gastrointest Endosc 2007; 65:832-41. 17. Fritcher EG, Halling KC. Advanced cytologic approaches for the diagnosis of pancreatobiliary cancer. Curr Opin Gastroenterol 2010;26:259-64. 18. Kim YS, Kim HG, Han J, et al. The significance of p53 and K-ras immunocytochemical staining in the diagnosis of malignant biliary obstruction by brush cytology during ERCP. Gut Liver 2010;4:219-25. 19. Papachristou GI, Smyrk TC, Baron TH. Endoscopic retrograde cholangiopancreatography tissue sampling: when and how? Clin Gastroenterol Hepatol 2007;5:783-90.

Availability of Journal back issues As a service to our subscribers, copies of back issues of Gastrointestinal Endoscopy for the preceding 5 years are maintained and are available for purchase from Elsevier until inventory is depleted. Please write to Elsevier Inc., Subscription Customer Service, 3251 Riverport Lane, Maryland Heights, MO 63043 or call 800-654-2452 or 314-447-8871 for information on availability of particular issues and prices.

804 GASTROINTESTINAL ENDOSCOPY Volume 75, No. 4 : 2012

www.giejournal.org