2812 Impact of Rapid On-Site Evaluation (ROSE) on Endoscopic Ultrasound-Guided FNA of Pancreatic Cysts: A Five-Year Comparative Study Between Two Academic Institutions

Abstract

INTRODUCTION: Endoscopic Ultrasound Guided-Fine Needle Aspiration (EUS-FNA) is considered the modality of choice for diagnosing intermediate pancreatic lesions. Rapid-on site evaluation (ROSE), when performed by a cytopathologist, has been suggested to improve the diagnostic yield of EUS-FNA. Prior studies were inconsistent regarding the benefit of ROSE. In our study, we evaluated the diagnostic yield of EUS-FNA in the presence of an on-site cytopathologist providing real-time feedback to the endoscopist compared to EUS-FNA alone in two academic hospitals. METHODS: We conducted a 2010-2015 retrospective chart review of all EUS-FNA performed at two academic medical centers (UT and UR). The policy at UT is to perform all EUS-FNA with ROSE, whereas it is not routinely done at UR. Cases where included when surgical pathology was available. Patients’ demographics, pancreatic cyst characteristics and EUS techniques were analyzed. EUS-FNA were considered valuable when it provided a classifying cytopathologic diagnosis. Student T-test and Analysis of Variance (ANOVA) were performed when appropriate for descriptive purposes. RESULTS: Forty-five cases (13 at UT, 32 at UR) were identified. EUS FNAs coupled with ROSE (UT) rendered 100% (13/13) valuable diagnoses, while those done with no cytopathologist on site (UR) had valuable diagnosis in 37.5% of the cases (12/32, P < 0.0001). The sensitivity of EUS-FNA at UT was 85% compared to 33% at UR. There was no statistically significant difference in gender, age, site or size of the tumor. The mean number of needle passes at UT was 1.9 (range 1-6) vs. 1.1 (range 1-3) at UR. The majority of needles used were 19G (55%) at UT vs 22G (83%) at UR (P < 0.001). Only one patient at UR had a minor cyst bleeding to the main pancreatic duct that required no therapeutic action. CONCLUSION: The results highlight the advantage of bed-side cytopathologist interaction with the endoscopist during EUS-FNA of pancreatic cysts. Absence of malignant cells despite suspicious ultrasound findings may have prompted more needle passes. The use of 19G needle may have improved outcomes with no significant increase of complications. In addition to the small sample size and the retrospective nature, the study is limited by the variability in personnel experience (endoscopists and cytopathologists) and the techniques used, indicating that further studies are necessary to confirm these findings.