OBJECTIVE: Our objective was to evaluate the accuracy of MDCT features of pancreatic cystic lesions in cyst characterization and in predicting cyst biologic aggressiveness. SUBJECTS AND METHODS: In this prospective study, 114 patients (40 men and 74 women; age range, 23-89 years) with 130 cystic lesions (size range, 31-160 mm) in the pancreas underwent contrast-enhanced dual-phase (n = 92) and portal phase (n = 22) examinations with 16- or 64-MDCT scanners. Using defined morphologic features of cystic lesions on MDCT, two readers performed blinded evaluations for cystic characterization and predicting biologic aggressiveness (invasive lesions, carcinoma in situ, and moderate grade dysplasias) before pancreatic surgery. Receiver operating characteristic analysis was performed to assess the accuracy of MDCT using pathologic evaluation of the surgical specimen as a reference standard. RESULTS: On the basis of MDCT features, the radiologic accuracy (reader 1 and reader 2) for stratifying lesions into mucinous and nonmucinous subtypes was 85% and 82% and for recognizing cysts with aggressive biology was 86% and 85%, respectively. Predictive values of MDCT were superior for lesions > 30 mm and nonmucinous lesions. Features favoring aggressive biology were main pancreatic duct dilation > 10 mm (p < 0.0001), biliary obstruction (p=0.01), mural nodule (p < 0.0001), main-duct intraductal papillary mucinous neoplasm (p < 0.0001), and advanced age (p = 0.0001). Sensitivity of detecting morphologic features was higher with the dual-phase pancreatic protocol CT. CONCLUSION: Morphologic features of pancreatic cystic lesions on MDCT allow reliable characterization into mucinous and nonmucinous subtypes and enable prediction of biologic aggressiveness.
OBJECTIVE: Our objective was to evaluate the accuracy of MDCT features of pancreatic cystic lesions in cyst characterization and in predicting cyst biologic aggressiveness. SUBJECTS AND METHODS: In this prospective study, 114 patients (40 men and 74 women; age range, 23-89 years) with 130 cystic lesions (size range, 31-160 mm) in the pancreas underwent contrast-enhanced dual-phase (n = 92) and portal phase (n = 22) examinations with 16- or 64-MDCT scanners. Using defined morphologic features of cystic lesions on MDCT, two readers performed blinded evaluations for cystic characterization and predicting biologic aggressiveness (invasive lesions, carcinoma in situ, and moderate grade dysplasias) before pancreatic surgery. Receiver operating characteristic analysis was performed to assess the accuracy of MDCT using pathologic evaluation of the surgical specimen as a reference standard. RESULTS: On the basis of MDCT features, the radiologic accuracy (reader 1 and reader 2) for stratifying lesions into mucinous and nonmucinous subtypes was 85% and 82% and for recognizing cysts with aggressive biology was 86% and 85%, respectively. Predictive values of MDCT were superior for lesions > 30 mm and nonmucinous lesions. Features favoring aggressive biology were main pancreatic duct dilation > 10 mm (p < 0.0001), biliary obstruction (p=0.01), mural nodule (p < 0.0001), main-duct intraductal papillary mucinous neoplasm (p < 0.0001), and advanced age (p = 0.0001). Sensitivity of detecting morphologic features was higher with the dual-phase pancreatic protocol CT. CONCLUSION: Morphologic features of pancreatic cystic lesions on MDCT allow reliable characterization into mucinous and nonmucinous subtypes and enable prediction of biologic aggressiveness.
Authors: Konstantin Dmitriev; Arie E Kaufman; Ammar A Javed; Ralph H Hruban; Elliot K Fishman; Anne Marie Lennon; Joel H Saltz Journal: Med Image Comput Comput Assist Interv Date: 2017-09-04
Authors: Richard K G Do; Seth S Katz; Marc J Gollub; Jian Li; Jennifer LaFemina; Emily C Zabor; Chaya S Moskowitz; David S Klimstra; Peter J Allen Journal: AJR Am J Roentgenol Date: 2014-11 Impact factor: 3.959
Authors: Dushyant V Sahani; Pietro A Bonaffini; Onofrio A Catalano; Alexander R Guimaraes; Michael A Blake Journal: Radiographics Date: 2012 Jul-Aug Impact factor: 5.333
Authors: Lawrence Mj Best; Vishal Rawji; Stephen P Pereira; Brian R Davidson; Kurinchi Selvan Gurusamy Journal: Cochrane Database Syst Rev Date: 2017-04-17