Background: Several changes have been made to the revised 2017 international consensus guidelines for management of pancreatic intraductal papillary mucinous neoplasms (IPMNs). However, the diagnostic performance is yet to be verified.
Purpose:To evaluate the revised guidelines for predicting malignant potential of pancreatic IPMNs and to compare diagnostic performance and intermodality agreement between contrast material-enhanced CT and MRI.
Materials and Methods:In this retrospective study, two radiologists analyzed the preoperative contrast-enhanced CT and MRI of patients with surgically resected pancreatic IPMNs from January 2007 to December 2017. The diagnostic performance of CT and MRI were analyzed by using receiver operating curve analysis. Intermodality agreement was assessed by using weighted k and intraclass correlation coefficient values.Results: A total of 86 patients (mean age, 67.6 years 6 8.9 [standard deviation]; 47 men and 39 women) with pancreatic IPMNs (benign, 58; malignant, 28) were included. At both CT and MRI, enhancing mural nodule (P , .001), abrupt main pancreatic duct caliber change (P , .001), lymphadenopathy (P = .006), larger main pancreatic duct size (P = .003), and faster cyst growth rate (P = .04) were more common in malignant than benign IPMNs. Irrespective of the modality, enhancing mural nodule of 5 mm or greater had the highest odds ratio (25 at CT vs 29 at MRI). The diagnostic performance of CT (area under the receiver operating characteristic curve, 0.83 [95% confidence interval: 0.75, 0.92]) and MRI (area under the receiver operating characteristic curve, 0.86 [95% confidence interval: 0.77, 0.95]) for predicting malignant IPMNs were comparable (P = .43), with good intermodality agreement (k = 0.70).
Conclusion:Among revised features, enhancing mural nodule of 5 mm or greater had the strongest association with malignant intraductal papillary mucinous neoplasm (IPMN), and diagnostic performance for prediction of malignant IPMNs were comparable between contrast-enhanced CT and MRI with good intermodality agreement.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.