Prediction of gallstone composition: Synthesis of CT and radiographic features in vitro

J. A. Brink, B. Kammer, P. R. Mueller, D. M. Balfe, E. L. Prien, J. T. Ferrucci

Research output: Contribution to journalArticle

20 Scopus citations

Abstract

PURPOSE: To test a morphoradiographic algorithm designed to predict the composition of gallstones with use of computed tomography (CT) to define calcification patterns. MATERIALS AND METHODS: Two reviewers retrospectively evaluated the radiographic features of 120 separate in vitro specimens (59 radiopaque and 61 radiolucent), then classified the stones into several categories of composition with the algorithm. RESULTS: The most useful features for prediction of cholesterol composition were, in order of decreasing importance, stone shape, absence of dense calcification on plain radiographs, overall CT attenuation not higher than that of water, presence of a second generation of smaller stones, and a peripheral cover of calcification detected on CT scans of aging cholesterol stones. The greatest error occurred in distinction between stones with mixed composition (50%-79% cholesterol) and older stones with higher cholesterol content (80%-95% cholesterol). CONCLUSION: The range of qualitative CT appearances advances the possibility of predicting gallstone composition and potential outcome of nonsurgical treatment.

Original languageEnglish
Pages (from-to)69-75
Number of pages7
JournalRadiology
Volume190
Issue number1
DOIs
StatePublished - Jan 1 1994

Keywords

  • Cholesterol
  • Computed tomography (CT), tissue characterization
  • Gallbladder, CT
  • Gallbladder, calculi
  • Gallbladder, radiography

Fingerprint Dive into the research topics of 'Prediction of gallstone composition: Synthesis of CT and radiographic features in vitro'. Together they form a unique fingerprint.

  • Cite this