Can Imaging Identify the Most Dangerous Pancreatic Cystic Neoplasms?
Endoscopic ultrasound can be used to identify cystic neoplasms of the pancreas that are most likely to become malignant, according to the February issue of Clinical Gastroenterology and Hepatology.
Mucus-producing cystic neoplasms of the pancreas, including intraductal papillary mucinous neoplasm (IPMN) and mucinous cystic neoplasm (MCN), that have mural nodules are most likely to become malignant and should be surgically removed. The nodules can be clearly identified by histology analysis, but Ning Zhong et al. investigated whether they could also be identified by imaging techniques such as computed tomography (CT) or endoscopic ultrasound (EUS).
Zhong et al. analyzed pathology specimens from 57 patients with resected branch duct (BD)-IPMNs or MCNs, along with preoperative CT and EUS images from most of the patients. They determined how frequently the imaging approaches accurately identified mural nodules, based on histologic analysis as the standard. They also looked for imaging features could distinguish mural nodules from mucus in pancreatic cysts.
Of the 57 patients, 22 were found to have a mural nodule, based on histologic analysis; 23% of the cysts that had nodules were malignant, compared to 3% of cysts without nodules. Altogether, 83% of cysts that contained invasive cancer or high-grade dysplasia also contained a mural nodule.
CT analysis identified nodules with 24% sensitivity and 100% specificity, compared with histology. CT did not identify mucus in cysts.
EUS identified mural nodules with 75% sensitivity and 83% specificity. Most of the echogenic lesions in cysts were mucus—not mural nodules. But EUS could distinguish cysts with nodules from those with mucus based on movement of the echogenic lesion when patients changed body positions or during fine-needle aspiration, and also by patterns of blood flow within the lesion.
Although there was only a low level of interobserver agreement among 10 endosonographers in diagnosis of nodules, they agreed on 3 sonographic features (echogenicity, edge and rim) that distinguished mucus from nodules (see figure).
In a validation set, the diagnostic accuracy of the sonographers improved from 57% to 79%, after they were educated about the 3 features of mural nodules. Diagnostic accuracy was 90% when all 3 features were present.
Zhong et al. say that these findings might refine the use of consensus criteria for resection of BD-IPMN by limiting the number of resections performed for mural nodules that are actually mucus.
They also state that because of its low level of sensitivity, CT is not the ideal method for screening or surveillance of patients with mucus-producing cystic neoplasms in whom identification of a mural nodule is likely to change management.
Nonetheless, EUS missed 3 nodules that were smaller than 5 mm (no intracystic echogenic lesion was observed). Furthermore mucus can overlay or mix with nodules—the authors were unable to distinguish cysts with mucus and nodules from those with nodules alone.
The authors conclude that EUS might become the preferred detection method for cysts with mural nodules, if specific criteria are used to distinguish nodules from mucus.
More Information on Pancreatic Cysts:
Read the article online.
Zhong N, Zhang L, Takahashi N, et al. Histologic and imaging features of mural nodules in mucinous pancreatic cysts. Clin Gastroenterol and Hepatol 2012;10:192–198.e2.