International consensus guidelines for surgical resection of mucinous neoplasms cannot be applied to all cystic lesions of the pancreas.

ABSTRACT

BACKGROUND & AIMS: International consensus guidelines, aimed at predicting malignancy, are available for surgical resection of mucinous cysts but not for other cystic lesions of the pancreas. We sought to determine whether the consensus guidelines can be applied to all cystic lesions of the pancreas. METHODS: We identified all patients who underwent surgical resection of pancreatic cysts from 2001-2007. Pathology analyses of surgical specimens served as the reference standard. Surgical resection criteria proposed by the Sendai Guidelines and 5 modifications of these criteria were tested to determine their accuracy for diagnosis of malignant cysts. RESULTS: Patients with cystic lesions of the pancreas (n = 154; mean age, 59.8 years; 64% women) underwent resection and met prespecified study criteria. Twenty-one patients had a malignancy. The classification cyst size > or = 3 cm had an accuracy of 56%, negative predictive value of 84%, and identified only 57% of the malignant cysts. The classification cyst size > or = 3 cm or cyst with main pancreatic duct > or = 10 mm had an accuracy of 55%, negative predictive value of 86%, and identified 66% of malignant cysts. The modified criterion of cyst size > or = 3 cm or cyst with main pancreatic duct > 3 mm had an accuracy of 48%, negative predictive value of 94%, and identified 91% (19/21) of the malignancies. Cyst size (odds ratio, 1.05) and pancreatic duct dilation > 3 mm (odds ratio, 10.5) were strong and independent predictors of malignancy. CONCLUSIONS: When applied to all cystic lesions of the pancreas, the international consensus criteria cause some malignant cysts to be missed. Modified criteria could identify most malignant cysts, although overall accuracy remains low.