Cystic pancreatic tumors: CT and sonographic assessment

C. D. Johnson, D. H. Stephens, J. W. Charboneau, H. A. Carpenter, T. J. Welch

Research output: Contribution to journalArticlepeer-review

202 Scopus citations


Thirty-five CT scans and 23 sonograms of 45 pathologically proved pancreatic cystic neoplasms (16 microcystic adenomas, 17 mucinous [macrocystic] cystadenomas, and 12 mucinous [macrocystic] cystadenocarcinomas) in 43 patients were retrospectively and blindly reviewed. Radiologic findings and their usefulness in differentiating microcystic from mucinous subtypes were assessed. The number of cysts within the tumors (more than six in microcystic adenomas and six or fewer in mucinous cystadenomas and cystadenocarcinomas) and the diameter of the majority of cysts within the tumor (≤2 cm in microcystic adenomas and >2 cm in mucinous tumors) were the most helpful radiologic findings in differentiating tumor types. Calcification was present in 38% of microcystic adenomas, 18% of mucinous cystadenomas, and 8% of mucinous cystadenocarcinomas by CT. Calcification was not definitely identified on any of the sonograms. A central scar was identified in only two (13%) of 16 microcystic adenomas. Blind retrospective review was often able to correctly classify tumors as either microcystic (CT, 93%; sonography, 78%) or mucinous (CT, 95%; sonography, 93%). All tumors with a typical appearance on either CT or sonography were categorized correctly. Cystic pancreatic tumors may be difficult to prospectively separate from other types of pancreatic cysts. Assuming a cystic neoplasm is present, it often can be classified correctly as microcystic or mucinous (macrocystic) by using the above criteria.

Original languageEnglish (US)
Pages (from-to)1133-1138
Number of pages6
JournalAmerican Journal of Roentgenology
Issue number6
StatePublished - Jan 1 1988

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Cystic pancreatic tumors: CT and sonographic assessment'. Together they form a unique fingerprint.

Cite this