BACKGROUND: Differentiation between benign and malignant branch-duct type intraductal papillary mucinous neoplasms (IPMNs) remains challenging. OBJECTIVE: To examine the usefulness of pancreatic duct lavage cytology with cell block method for discriminating benign and malignant branch-duct type IPMNs. PATIENTS AND METHODS: Between December 2007 and April 2011, patients with branch-duct type IPMNs having mural nodules on EUS were examined by pancreatic duct lavage cytology by using the cell block method. Cell block sections underwent hematoxylin and eosin staining and mucin immunostainings (MUCs 1, 2, 5AC, and 6). DESIGN: Single-center, prospective study. SETTING: Academic medical center. MAIN OUTCOME MEASUREMENTS: The sensitivity and specificity of cytology were assessed. The agreement between cytological and histological results for MUC was also examined. RESULTS: Cytology with this method was investigated in 44 patients. Cell block diagnosis was cancer positive (class V or IV) in 11 patients and negative (classes I, II, III, and noninformative) in 33. The sensitivity, specificity, and positive and negative predictive values of this method were 92%, 100%, 100%, and 97%, respectively. The cytological and histological results of MUCs 1, 2, 5AC, and 6 agreed in 88% (15/17), 94% (16/17), 88% (15/17), and 100% (17/17), respectively. LIMITATIONS: Single center and small number of patients. CONCLUSIONS: Pancreatic duct lavage cytology with the cell block method may be useful to differentiate between benign and malignant IPMNs preoperatively and as well as to determine their mucin type.
BACKGROUND: Differentiation between benign and malignant branch-duct type intraductal papillary mucinous neoplasms (IPMNs) remains challenging. OBJECTIVE: To examine the usefulness of pancreatic duct lavage cytology with cell block method for discriminating benign and malignant branch-duct type IPMNs. PATIENTS AND METHODS: Between December 2007 and April 2011, patients with branch-duct type IPMNs having mural nodules on EUS were examined by pancreatic duct lavage cytology by using the cell block method. Cell block sections underwent hematoxylin and eosin staining and mucin immunostainings (MUCs 1, 2, 5AC, and 6). DESIGN: Single-center, prospective study. SETTING: Academic medical center. MAIN OUTCOME MEASUREMENTS: The sensitivity and specificity of cytology were assessed. The agreement between cytological and histological results for MUC was also examined. RESULTS: Cytology with this method was investigated in 44 patients. Cell block diagnosis was cancer positive (class V or IV) in 11 patients and negative (classes I, II, III, and noninformative) in 33. The sensitivity, specificity, and positive and negative predictive values of this method were 92%, 100%, 100%, and 97%, respectively. The cytological and histological results of MUCs 1, 2, 5AC, and 6 agreed in 88% (15/17), 94% (16/17), 88% (15/17), and 100% (17/17), respectively. LIMITATIONS: Single center and small number of patients. CONCLUSIONS:Pancreatic duct lavage cytology with the cell block method may be useful to differentiate between benign and malignant IPMNs preoperatively and as well as to determine their mucin type.