Mo1413 Accuracy of Ex Vivo Imaging of the Liver and Pancreas Using a High Resolution Microendoscope (HRME)

uncinate process (1). Cytology results on initial FNA were reported as benign in 15(71.4%), cellular atypia in 2(9.5%), inadequate in 3(14.3%), and suspicious in 1(4.8%). Among patients with non-cystic lesions (13), malignancy was confirmed in 9/13(69.2%) cases including adenocarcinoma (7), lymphoma (1), and pseudopapillary tumor of pancreas (1). One patient had negative cytology on repeat EUS-FNA but was confirmed to have adenocarcinoma on CT-FNA. 3/14 patients with noncystic lesion were confirmed to have benign etiologies on repeat EUSFNA, including chronic pancreatitis in 2 with no evidence of malignancy at a follow up of 30.7 and 10.9 months respectively; and an indeterminate biliary stricture in 1, which resolved on clinical follow up of 20.2 months. In patients with cystic lesions of pancreas (8), IPMN was confirmed in 2, MCN in 1, serous cystadenoma in 3, and no significant pathology in 2 cases. One case of IPMN was found to have dysplasia on repeat EUS-FNA. All other lesions were determined to have benign course based on a median follow up of 51.0 months (13.3-85). Repeat EUS altered clinical decision making in 9/13(69.2%) cases with non-cystic lesions as compared to 0/8 cases with cystic lesions (p-value 0.01; Fisher’s exact test). The diagnostic accuracy of repeat EUS-FNA in identifying malignancy or a potentially malignant lesion was 90.9%. Conclusion: Repeat EUSFNA alters clinical decision making in majority of non-cystic lesions of pancreas. However, repeat EUS-FNA may not be as helpful in cases of cystic lesions in this setting.