It is known that most IPMNs of the branch-duct this website type are less invasive and can be followed4, 5 and 6; thus, differentiation between benign and malignant tumors must be accurate to indicate surgical resection. We have already demonstrated that pancreatic duct lavage cytology is of high diagnostic accuracy because it allows the accumulation of a sufficient number of neoplastic cells exfoliated from the branch pancreatic duct.7 In this study, we examined the usefulness of pancreatic duct lavage

cytology with the cell block method for discriminating benign IPMNs of the branch-duct type from malignant ones. The cell block method allows cytological and/or histological evaluation with hematoxylin and eosin (H&E) stain and with mucin immunostaining (MUC) (MUC1, 2, 5AC, and 6).8 Mucins are high molecular weight

glycoproteins,9 and the malignant potential of IPMNs is reported to differ depending on their mucin type characterized by the MUC.10 and 11 Between December 2007 and April 2011, patients in our outpatient clinic who were suspected of having branch-duct type IPMNs by CT or magnetic resonance imaging (MRI) underwent EUS, and patients having mural nodules on EUS were examined by endoscopic retrograde pancreatography (ERP) followed by pancreatic duct lavage cytology. MRI/CT findings as indicators of branch-duct type IPMNs appear as clusters PLX3397 of small cysts with a grapelike appearance or as a single cystic lesion with lobulated or irregular margins and sparse septa, often with dilation of the pancreatic duct near the lesion.12 A mural nodule in this study was defined as an EUS-detectable echogenic protruding component in an ectatic branch pancreatic duct (Fig. 1). The diagnosis was confirmed based on the presence mafosfamide of abnormally dilated branch pancreatic ducts accompanied by intraductal mucin on ERP. Intraductal mucin was detected as a mobile and

amorphous filling defect in the pancreatic duct. The type of IPMN was determined according to the World Health Organization classification.13 Surgical intervention was indicated when the results of cytology were positive, or when mural nodules larger than 5 mm or a pancreatic mass was detected by EUS. Patients with no indications for surgery were followed for more than 12 months, during which thin-slice (1-2 mm) CT or MRI with contrast enhancement was performed every 3 to 4 months. Patients who showed progressive enlargement of the main and the ectatic branch pancreatic ducts, mural nodules, or a pancreatic mass during follow-up on CT or MRI underwent EUS, and surgery was indicated when mural nodules larger than 5 mm or a pancreatic mass was detected by EUS.