Rapid On-Site Evaluation for Endoscopic Ultrasound- Guided Fine-Needle Biopsy of the Pancreas Decreases the Incidence of Repeat Biopsy Procedures

Rapid On-Site Evaluation for Endoscopic Ultrasound- Guided Fine-Needle Biopsy of the Pancreas Decreases the Incidence of Repeat Biopsy Procedures Brian T. Collins, MD 1 ; Faris M. Murad, MD 2 ; Jeff F. Wang, MD 1 ; and Cory T. Bernadt, MD, PhD 1 BACKGROUND: Rapid on-site evaluation (ROSE) for endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) biopsy of the pancreas provides immediate feedback regarding cellular adequacy to aid in obtaining a definitive diagnosis and has the potential to avoid repeat procedures. The objective of the current study was to measure the impact of ROSE service on the incidence of repeat EUS FNA biopsy procedures. METHODS: Over a consecutive 3-year period, the pathology database at Washington University Medical Center was searched for patients with both an initial and subsequent EUS FNA biopsy demonstrating a solid lesion of the pancreas. These were divided temporally between the time before and after the introduction of ROSE service. Reports were reviewed and results were recorded. RESULTS: A total of 379 patients underwent ROSE service and 377 patients did not. The percentage of repeat non-rose EUS FNA cases was 5.8% and the percentage of repeat ROSE EUS FNA cases was 2.9%. The use of the ROSE service was found to decrease the number of repeat procedures by approximately 50% (P 5.024). For those patients who underwent a repeat EUS-FNA procedure, the ROSE service provided a higher rate of definitive diagnosis among patients undergoing repeat procedures (67%) versus the non-rose cohort (27%). CONCLUSIONS: The use of ROSE for EUS-FNA biopsy of the pancreas was found to result in fewer patients undergoing repeat procedures. Patients who required a repeat procedure with the use of ROSE had a higher percentage of definitive diagnostic categorization on the repeat biopsy. Initial use of ROSE for EUS- FNA of solid pancreatic lesions was found to decrease the number of patients who required a repeat procedure. Cancer (Cancer Cytopathol) 2013;121:518 24. VC 2013 American Cancer Society. KEY WORDS: rapid on-site evaluation (ROSE), fine-needle aspiration, endoscopic ultrasound, pancreas, neoplasm. INTRODUCTION Endoscopic ultrasound (EUS) along with fine-needle aspiration (FNA) is the primary technique used to evaluate patients suspected of having pancreatic lesions as well as to acquire tissue. EUS involves the use of an echoendoscope to sonographically observe organs adjacent to the gastrointestinal tract and perform FNA biopsies of lesions of interest. EUS of the pancreas allows for detailed sonographic evaluation of the complete gland and ductal anatomy. The pathologic entities encountered in the pancreas can range from abnormalities of the pancreatic duct, cystic lesions, and solid masses. The endosonographer is able to sonographically observe areas of interest within the pancreas and perform FNA under sonographic guidance. Rapid on-site evaluation (ROSE) performed by the cytopathologist is necessary to direct the endosonographer as to whether the aspirate obtained is sufficient to permit a definitive diagnosis. The ROSE process involves the evaluation of the direct smears obtained at the point of care in the endoscopy suite. The direct smears are quickly processed and examined by a Corresponding author: Brian T. Collins, MD, Department of Pathology and Immunology, Washington University School of Medicine, Campus Box 8118, 660 South Euclid Ave, St. Louis, MO 63110-1093; Fax: (314) 747-2663; bcollins@path.wustl.edu 1 Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri; 2 Division of Gastroenterology and Interventional Endoscopy, Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri Received: May 17, 2013; Revised: July 1, 2013; Accepted: July 9, 2013 Published online August 23, 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cncy.21340, wileyonlinelibrary.com 518 Cancer Cytopathology September 2013

ROSE for EUS FNA in Repeat Procedures/Collins et al light microscope in the endoscopy procedure suite with immediate direct feedback provided to the performing endosonographer. This information can assist in guiding the number of EUS-FNA passes required to obtain a final diagnosis. This immediate evaluation assists the endosonographer in knowing whether the aspirates obtained are diagnostic or nondiagnostic. When they are nondiagnostic, the biopsy technique can be adjusted or additional EUS-FNA passes made in an attempt to obtain cytological diagnosis. 1 When EUS-FNA is performed without ROSE, the endosonographer does not have the pathologic information provided by an immediate assessment, and is unaware if the aspirates obtained are adequate to yield a definitive diagnosis. Solid masses are typically subject to the immediate on-site evaluation, whereas cystic lesions are not. The ROSE service was introduced for EUS-FNA biopsy procedures in our institution in 2011. The objective of the current study was to measure the impact of ROSE service on the incidence of repeat EUS-FNA biopsy procedures for patients with solid masses of the pancreas. MATERIALS AND METHODS The introduction of ROSE for the EUS service was identified as the midpoint of the patient-matched case-controlled study. The pathology database at Washington University Medical Center was searched for cases undergoing EUS-FNA before and after the introduction of the ROSE service. The post-eus ROSE service was collected for 18 consecutive calendar months (1.5 years) and was initiated on July 1, 2011. A case-controlled matched cohort of patients predating the introduction of the ROSE service was collected for 18 consecutive calendar months (1.5 years). These occurred with a similar patient cohort mix and within a single practice setting. This time span was judged to be adequate for sufficient case volumes to produce meaningful data. The Institutional Review Board approved the study. Attending endoscopists, with the assistance of trainees, performed all EUS-FNA cases. predating ROSE had a member of the cytology laboratory present (cytotechnologist) to prepare smears and process the FNA samples. They did not have a microscope or a pathologist present in the endoscopy suite to examine the case. The process in the endoscopy procedure suite included the initial FNA material expressed on a slide, which was split=paired into 2 slides: one was air-dried for modified Wright-Giemsa style staining and the other was placed in 95% alcohol fixation for Papanicolaou staining, both of which were performed after the procedure in the cytopathology laboratory. Needle rinse material was collected in RPMI cell preservative and processed for cell block if there was adequate material per site or for alcohol-fixed Papanicolaou-stained cell concentration slides (cytospin) if the amount of material was insufficient for cell block. Needles were rinsed after the preparation of direct aspirate smears and dedicated aspirate samples were made for cell block preparations. Slides were numbered sequentially (1, 2, etc) to designate pass numbers and to indicate different anatomic sites when necessary. postdating the implementation of the ROSE service included the cytotechnologist who processed the FNA and the attending cytopathologist with trainee(s) (either fellow and=or core pathology resident[s]). The FNA material was handled the same as for cases obtained before ROSE except that the air-dried slides were stained by modified Wright-Giemsa staining in the endoscopy suite. The cytopathology team would examine the airdried, modified, Wright-Giemsa stained smears in the endoscopy suite at the time of the procedure, and an immediate evaluation was communicated directly to the endosonographer performing the EUS-FNA procedure. The service was provided at multiple evaluation episodes during an FNA biopsy. The pathologist who attended the immediate evaluation was assigned the case for sign out. Once the ROSE service was introduced for the EUS-FNA procedures, all cases used the on-site immediate evaluation service. The pathology reports for cases of EUS-FNA were collected before the implementation of the ROSE service and for EUS-FNA procedures for 18 consecutive months after the introduction of the ROSE service. The reports were reviewed and data collected. The data collected included the number of patients, their sex, and age; the anatomic sites biopsied; the pathologist of record; whether ROSE was performed or not performed; whether the mass was solid or cystic on imaging; the number of direct slide smears for each anatomic site; the time between the initial and second biopsy; and the final diagnostic categories for the initial and second biopsy. The general diagnostic categories used were unsatisfactory, negative for malignancy, atypical, suspicious, and positive for malignancy. Correlation between the ROSE interpretation and Cancer Cytopathology September 2013 519

the final case interpretation (whether the initial or second biopsy) was not undertaken. Statistical Analysis The data was analyzed using SPSS Statistical software (version 21; SPSS Inc, Chicago, Ill) with independent t- square and significance of the differences between 2 independent percentages calculated for P values. A P value.05 was considered to be statistically significant. RESULTS There were 379 consecutive patients identified in the designated time-length database search after the introduction of ROSE EUS-FNA (18 months). A matched casecontrolled EUS-FNA cohort of 377 patients was identified in the preceding 18 months before the introduction of the ROSE service. All patients included had solid lesions of the pancreas; patients with cystic lesions were excluded from the study. Clinical and pathologic features are presented in Table 1. The non-rose cohort had a male:female ratio of 14:8, with average age of 57.9 years. Seventeen of the 22 cases were obtained from the head of pancreas, with an average of 6.3 weeks between biopsies. Seventeen of these 22 cases had a different pathologist report the final interpretation of the second biopsy. The ROSE cohort had a male:female ratio of 3:8 with average age of 65.8 years. The distribution of anatomic sites in the pancreas was even, with an average of 7.1 weeks between biopsies; 8 of 11 cases had a different pathologist report the final interpretation of the second biopsy (Table 1). Within the non-rose cohort, there were 22 repeat EUS-FNA biopsies performed, whereas within the ROSE cohort, there were 11 repeat EUS-FNA biopsies. This represents twice the incidence in the non-rose cohort compared with the ROSE cohort. This represented a proportion of 0.0584 for the non-rose cohort and 0.029 for the ROSE group, with a difference of 0.0293 (P <.024) (Table 2). With the repeat cases, all subsequent EUS-FNA results were classified for the non-rose and ROSE patients. For the 22 non-rose repeat cases, the initial diagnostic categories included 1 unsatisfactory (5%), 10 benign (45%), 8 atypical (36%), and 3 suspicious (14%) findings. For the 22 non-rose repeat cases, the second biopsy diagnostic categories included 0 unsatisfactory TABLE 1. ROSE EUS FNA Biopsy in Repeat Procedures: Clinical Information a Non-ROSE Data Total no. of patients 377 No. of repeat EUS FNA biopsies 22 Sex Male 14 Female 8 Age (average), y 57.9 Site of biopsy Head 17 Neck 1 Uncinate 2 Body 2 Tail 0 Second biopsy pathologist different 17 Second biopsy pathologist same 5 Time between initial and second biopsy, wk 6.3 ROSE Data Total no. of patients 379 No. of repeat EUS FNA biopsies 11 Sex Male 3 Female 8 Age (average), y 65.8 Site of biopsy Head 4 Neck 2 Uncinate 0 Body 3 Tail 2 Second biopsy pathologist different 8 Second biopsy pathologist same 3 Time between initial and second biopsy, wk 7.1 a All lesions were found to be solid on imaging. TABLE 2. ROSE EUS FNA Biopsy in Repeat Procedures: Proportional Difference Biopsy Repeat Patients/ All Patients Proportional Ratio Non-ROSE service 22/377 0.0584 ROSE service 11/379 0.029 Difference 50% difference 20.0293 (P value <.024) (0%), 9 benign (41%), 5 atypical (23%), 2 suspicious (9%), and 6 positive (27%) findings (Table 3). For the 11 ROSE repeat cases, the initial diagnostic categories included 1 unsatisfactory (9%), 3 benign (27%), 5 atypical (45%), and 2 suspicious (18%) categories. For the 11 ROSE repeat cases, the second biopsy diagnostic categories included 0 unsatisfactory (0%), 1 benign (9%), 2 atypical (18%), 1 suspicious (9%), and 7 positive (64%) findings (Table 3). 520 Cancer Cytopathology September 2013

ROSE for EUS FNA in Repeat Procedures/Collins et al TABLE 3. ROSE EUS FNA Biopsy in Repeat Procedures: Diagnostic Categories for Initial and Second EUS FNA Biopsy TABLE 5. Non-ROSE EUS FNA Biopsy: Initial Category With Breakdown of Second Biopsy Categories Non-ROSE Patients (n 5 22) Initial Diagnostic Category Second Diagnostic Category Initial Biopsy Category (n 5 22) Second Biopsy Category Groups No. of Diagnostic category Unsatisfactory 1 (5%) 0 Benign 10 (45%) 9 (41%) Atypical 8 (36%) 5 (23%) Suspicious 3 (14%) 2 (9%) Positive 0 6 (27%) Total 22 22 ROSE Patients (n 5 11) Diagnostic category Unsatisfactory 1 (9%) 0 Benign 3 (27%) 1 (9%) Atypical 5 (45%) 2 (18%) Suspicious 2 (18%) 1 (9%) Positive 0 7 (64%) Total 11 11 TABLE 4. ROSE EUS FNA Biopsy in Repeat Procedures: Definitive Categorization After Second Biopsy Biopsy Definitive Diagnosis on Second Biopsy/All Patients Proportional Ratio Non-ROSE service 6/22 (27%) 0.273 ROSE service 7/11 (64%) 0.636 Difference 37% higher rate of positivity on ROSE second biopsy 0.364 (P value <.044) Twice as likely to have a definitive positive using ROSE than non- ROSE service Among the 22 patients in the non-rose cohort, 6 patients had a definitive diagnosis (positive) made on the second biopsy (27%). The non-rose cohort did not have the ROSE service performed at either the time of the first biopsy or for the second repeat biopsy. For the 11 patients in the ROSE cohort, 7 patients had a definitive diagnosis (positive) made on the second biopsy (64%). The proportional difference was 0.364 (P <.044). The ROSE cohort had the ROSE service performed at the time of first biopsy and for the second repeat biopsy (Table 4). For the non-rose cohort, the initial EUS-FNA biopsies demonstrated a mixed of unsatisfactory (1 case), benign (10 cases), atypical (8 cases), and suspicious (3 Unsatisfactory 1 Benign 1 (100%) Benign 10 Benign 6 (60%) Atypical 3 (30%) Positive 1 (10%) Atypical 8 Benign 1 (13%) Atypical 2 (25%) Suspicious 2 (25%) Positive 3 (37%) Suspicious 3 Benign 1 (33%) Positive 2 (66%) Total 22 22 TABLE 6. ROSE EUS FNA Biopsy: Initial Category With Breakdown of Second Biopsy Categories Initial Biopsy Category (n 5 11) No. of Second Biopsy Category Groups No. of Unsatisfactory 1 Positive 1 (100%) Benign 3 Positive 2 (66%) Benign 1 (33%) Atypical 5 Atypical 1 (20%) Suspicious 1 (20%) Positive 3 (60%) Suspicious 2 Positive 2 (100%) Total 11 11 cases) findings. These were repeated without the ROSE service and 6 of the 10 benign cases remained benign, with 3 converting to atypical and 1 converting to positive on second biopsy. The 8 atypical cases converted to 1 benign, 2 atypical, 2 suspicious, and 3 positive on second biopsy. The 3 suspicious cases converted to 1 benign and 2 positive. The 1 unsatisfactory case converted to benign (Table 5). For the ROSE cohort, the initial EUS-FNA biopsies showed a spectrum of unsatisfactory (1 case), benign (3 cases), atypical (5 cases), and suspicious (2 cases) findings. These were repeated with the ROSE service and 2 of the 3 benign category cases converted to positive. The other benign category case was in patient with chronic pancreatitis and was classified as benign on the second biopsy. The 5 atypical categories converted to atypical (1 case), suspicious (1 case), and positive (3 cases). The 2 suspicious cases were converted to positive, as was the 1 unsatisfactory case (Table 6), The initial benign category cases Cancer Cytopathology September 2013 521

in the non-rose and ROSE cohorts underwent repeat biopsy based on a high clinical index of suspicion for malignancy. DISCUSSION In our experience, the use of the ROSE service for EUS- FNA biopsy of solid masses in the pancreas resulted in fewer patients who required a second repeat procedure in comparison with those who did not have ROSE service. Among the 2 cohorts, the non-rose group had twice as many repeat EUS-FNA biopsies (22 biopsies vs 11 biopsies). The proportion difference (0.0293) was statistically significant (P <.024). There are a variety of factors that can potentially explain the difference. ROSE service is an active, dynamic, patient care activity that is performed at the point of care and provides real-time data and feedback to the performing endosonographer. When the endosonographer performs the first set of biopsies and is provided with the on-site immediate interpretation, the initial interpretation of adequacy is crucial to determine whether more aspirates are required or there is enough material for a definitive diagnosis. If the material is adequate for diagnosis, then the procedure can be concluded. However, if it is unsatisfactory or does not indicate a high likelihood of providing a definitive diagnosis, the endosonographer can use that information to make adjustments or obtain additional passes for assessment. Certain solid neoplasms of the pancreas are desmoplastic and paucicellular. These types of tumors may only provide a few atypical groups that are not sufficient to provide a definitive diagnosis. ROSE service is particularly helpful in these cases so that more aspirates are obtained to increase the yield of a definitive diagnosis. Because the slides are numbered, those with more cellular material can help the cytologist to direct the endosonographer in targeting the area of the lesion that yielded the most cells. Harris and Buscaglia advocated the use of ROSE and a variety of factors to increase diagnostic yield, including sampling the lesion in multiple planes, targeting the margins, or firmer ends of necrotic lesions. 1 The use of ROSE for the EUS-FNA biopsy of solid masses resulted in fewer repeat biopsies, and the patients in the ROSE cohort who underwent a second biopsy had a higher rate of definitive diagnoses on the second biopsy (63% for the ROSE cohort vs 27% for the non-rose cohort). The proportional difference was 0.364 (P <.044) and was statistically significant. Therefore, when the decision was made to perform a repeat EUS- FNA biopsy in both cohorts, the ROSE cohort was more than twice as more likely to provide a definitive positive classification. There are a variety of factors that can potentially explain this difference. The ROSE service can has the potential to provide the endosonographer with the assurance that they performed the most efficient initial biopsy possible, and therefore other subsequent clinical management choices are made based on this knowledge (close follow-up or surgical intervention instead of repeat EUS-FNA biopsy). Of course, ROSE service for a EUS-FNA biopsy of a solid mass of the pancreas does not provide an assurance of a definitive diagnosis. There are many factors that contribute to indeterminate categorization, including anatomic considerations, the location and nature of the mass=lesion, the endoscopic technique used, and the postbiopsy cytopathologic processing technique. To our knowledge, there are no studies published to date that examine the use of ROSE and its impact on repeat EUS-FNA. Tadic et al reported the use of repeat EUS-FNA and found it to be effective in providing a diagnosis; however, they did not use ROSE. 2 There are a variety of studies that have examined the use of ROSE for EUS-FNA biopsy of solid masses of the pancreas and other anatomic locations. A variety of these studies have demonstrated an advantage to using the ROSE service in EUS-FNA of the pancreas. Schmidt et al reported that ROSE provided a statistically significant improvement in adequacy rates for pancreatic lesions when performing a systematic review and metaanalysis of studies reporting the adequacy rate with and without ROSE. 3 Klapman et al found that the ROSE service improved the diagnostic yield of EUS-FNA, 4 and Jhala et al described the specific advantages of EUS-FNA ROSE. 5 These and other studies have described the advantages of ROSE in EUS-FNA of the pancreas. 1,6 8 The lack of ROSE has been reported to be associated with low EUS-FNA sensitivity. 9 To the best of our knowledge, none of the studies published to date that examined the use of ROSE versus not using ROSE for EUS-FNA of the pancreas examined its impact on second procedures. 4,10 13 In the current study, the endoscopists were an established group with long-standing experience with EUS-FNA. When the ROSE service is introduced to endoscopists, it is possible that as they gain 522 Cancer Cytopathology September 2013

ROSE for EUS FNA in Repeat Procedures/Collins et al experience with ROSE it could have a beneficial impact on the rate of second biopsies being performed. A study with more cases would have larger overall numbers; however, we believe that the trend and statistically valid difference in the 2 cohorts would persist. Studies of the use of ROSE at other anatomic locations outside the pancreas have shown an improvement in adequacy rates. 14 21 The correlation between the immediate interpretation and the final interpretation has been shown to be highly concordant. 22,23 In patients undergoing pulmonary endobronchial ultrasound-guided FNA biopsy, the useofrosedecreasedthecomplicationrateassociated with the necessary diagnostic procedures. 24,25 Nevertheless, some studies have demonstrated no clear advantage of the ROSE service in biopsies of other anatomic locations. 26 There are a variety of advantages and challenges when providing a ROSE service. 27 Relatively low reimbursement rates for the pathologist remains a problem. 28,29 Also, not all practice sites have access to the personnel and expertise required for immediate on-site evaluation. In circumstances in which access to cytopathologists is limited, the use of telepathology has been described for EUS-FNA ROSE. 30,31 In the current study, we examined 377 patients over an 18-month period who had undergone EUS-FNA biopsy of a solid pancreatic mass and a second biopsy without ROSE service. We examined 379 patients over an 18-month period who had undergone EUS-FNA biopsy of a solid pancreatic mass and a second biopsy with ROSE service. In the current study, use of the ROSE service for EUS-FNA biopsy of solid masses resulted in fewer second biopsies (proportional difference of 0.0293; P <.024), and the second biopsies in the ROSE cohort had a higher rate of definitive diagnosis (63%) compared with the non- ROSE cohort (27%). The use of ROSE in EUS-FNA of solid pancreatic lesions was found to decrease the number of repeat procedures due to nondiagnostic sampling without ROSE. The use of ROSE during EUS-FNA of solid pancreatic lesions appears to increase the possibility of obtaining a definitive diagnosis. FUNDING SUPPORT No specific funding was disclosed. CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures. REFERENCES 1. Harris MD, Buscaglia JM. How to do pancreatic mass FNA. Gastrointest Endosc. 2010;71:825-826. 2. Tadic M, Kujundzic M, Stoos-Veic T, Kaic G, Vukelic-Markovic M. Role of repeated endoscopic ultrasound-guided fine needle aspiration in small solid pancreatic masses with previous indeterminate and negative cytological findings. Dig Dis. 2008;26:377-382. 3. Schmidt RL, Witt BL, Matynia AP, Barraza G, Layfield LJ, Adler DG. Rapid on-site evaluation increases endoscopic ultrasoundguided fine-needle aspiration adequacy for pancreatic lesions. Dig Dis Sci. 2012;58:872-882. 4. Klapman JB, Logrono R, Dye CE, Waxman I. Clinical impact of onsite cytopathology interpretation on endoscopic ultrasound-guided fine needle aspiration. Am J Gastroenterol. 2003;98:1289-1294. 5. Jhala NC, Jhala DN, Chhieng DC, Eloubeidi MA, Eltoum IA. Endoscopic ultrasound-guided fine-needle aspiration. A cytopathologist s perspective. Am J Clin Pathol. 2003;120:351-367. 6. Chen J, Yang R, Lu Y, Xia Y, Zhou H. Diagnostic accuracy of endoscopic ultrasound-guided fine-needle aspiration for solid pancreatic lesion: a systematic review. J Cancer Res Clin Oncol. 2012; 138:1433-1441. 7. Hayashi T, Ishiwatari H, Yoshida M, et al. Rapid on-site evaluation by endosonographer during endoscopic ultrasound-guided fine needle aspiration for pancreatic solid masses. J Gastroenterol Hepatol. 2013;28:656-663. 8. Gimeno-Garcia AZ, Elwassief A. How to improve the success of endoscopic ultrasound guided fine needle aspiration cytology in the diagnosis of pancreatic lesions. J Interv Gastroenterol. 2012;2:31-36. 9. Dumonceau JM, Koessler T, van Hooft JE, Fockens P. Endoscopic ultrasonography-guided fine needle aspiration: relatively low sensitivity in the endosonographer population. World J Gastroenterol. 2012;18:2357-2363. 10. Iglesias-Garcia J, Dominguez-Munoz JE, Abdulkader I, et al. Influence of on-site cytopathology evaluation on the diagnostic accuracy of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) of solid pancreatic masses. Am J Gastroenterol. 2011;106:1705-1710. 11. Cleveland P, Gill KR, Coe SG, et al. An evaluation of risk factors for inadequate cytology in EUS-guided FNA of pancreatic tumors and lymph nodes. Gastrointest Endosc. 2010;71:1194-1199. 12. Cermak TS, Wang B, DeBrito P, Carroll J, Haddad N, Sidawy MK. Does on-site adequacy evaluation reduce the nondiagnostic rate in endoscopic ultrasound-guided fine-needle aspiration of pancreatic lesions? Cancer (Cancer Cytopathol). 2012;120:319-325. 13. Alsohaibani F, Girgis S, Sandha GS. Does onsite cytotechnology evaluation improve the accuracy of endoscopic ultrasound-guided fine-needle aspiration biopsy? Can J Gastroenterol. 2009;23:26-30. 14. Witt BL, Schmidt RL. Rapid on-site evaluation (ROSE) improves the adequacy of fine needle aspiration for thyroid lesions: a systematic review and meta-analysis. Thyroid. 2012;23:428-435. 15. Wohlschlager J, Darwiche K, Ting S, et al. Rapid on-site evaluation (ROSE) in cytological diagnostics of pulmonary and mediastinal diseases [in German]. Pathologe. 2012;33:308-315. 16. Fassina A, Corradin M, Zardo D, Cappellesso R, Corbetti F, Fassan M. Role and accuracy of rapid on-site evaluation of CTguided fine needle aspiration cytology of lung nodules. Cytopathology. 2011;22:306-312. 17. Tournoy KG, Praet MM, Van Maele G, Van Meerbeeck JP. Esophageal endoscopic ultrasound with fine-needle aspiration with an on-site cytopathologist: high accuracy for the diagnosis of mediastinal lymphadenopathy. Chest. 2005;128:3004-3009. 18. Diacon AH, Schuurmans MM, Theron J, et al. Utility of rapid on-site evaluation of transbronchial needle aspirates. Respiration. 2005;72:182-188. 19. Davenport RD. Rapid on-site evaluation of transbronchial aspirates. Chest. 1990;98:59-61. Cancer Cytopathology September 2013 523

20. Diette GB, White P Jr, Terry P, Jenckes M, Rosenthal D, Rubin HR. Utility of on-site cytopathology assessment for bronchoscopic evaluation of lung masses and adenopathy. Chest. 2000;117:1186-1190. 21. Chin R Jr, McCain TW, Lucia MA, et al. Transbronchial needle aspiration in diagnosing and staging lung cancer: how many aspirates are needed? Am J Respir Crit Care Med. 2002;166:377-381. 22. Factor R, Layfield LJ. Intraprocedural evaluation of fine-needle aspiration smears: how good are we? Diagn Cytopathol. 2011;40:760-763. 23. Woon C, Bardales RH, Stanley MW, Stelow EB. Rapid assessment of fine needle aspiration and the final diagnosis how often and why the diagnoses are changed. Cytojournal. 2006;3:25. 24. Eapen GA, Shah AM, Lei X, et al;american College of Chest Physicians Quality Improvement Registry, Education. Complications, consequences, and practice patterns of endobronchial ultrasound-guided transbronchial needle aspiration: results of the AQuIRE registry. Chest. 2013;143:1044-1053. 25. Trisolini R, Cancellieri A, Tinelli C, et al. Rapid on-site evaluation of transbronchial aspirates in the diagnosis of hilar and mediastinal adenopathy: a randomized trial. Chest. 2011;139:395-401. 26. Griffin AC, Schwartz LE, Baloch ZW. Utility of on-site evaluation of endobronchial ultrasound-guided transbronchial needle aspiration specimens. Cytojournal. 2011;8:20. 27. Gupta PK. Progression from on-site to point-of-care fine needle aspiration service: opportunities and challenges. Cytojournal. 2010; 7:6. doi: 10.4103/1742-6413.63195. 28. Layfield LJ, Bentz JS, Gopez EV. Immediate on-site interpretation of fine-needle aspiration smears: a cost and compensation analysis. Cancer. 2001;93:319-322. 29. Al-Abbadi MA, Bloom LI, Fatheree LA, et al. Adequate reimbursement is crucial to support cost-effective rapid on-site cytopathology evaluations. Cytojournal. 2010;7:22. 30. Buxbaum JL, Eloubeidi MA, Lane CJ, et al. Dynamic telecytology compares favorably to rapid onsite evaluation of endoscopic ultrasound fine needle aspirates. Dig Dis Sci. 2012;57:3092-3097. 31. Marotti JD, Johncox V, Ng D, Gonzalez JL, Padmanabhan V. Implementation of telecytology for immediate assessment of endoscopic ultrasound-guided fine-needle aspirations compared to conventional on-site evaluation: analysis of 240 consecutive cases. Acta Cytol. 2012;56:548-553. 524 Cancer Cytopathology September 2013