231 Michael J. Foley, M.D. Gary G. Ghahremani, M.D. Lee F. Rogers, M.D. Reappraisal of Contrast Media Used To Detect Upper Gastrointestinal Perforations Comparison of Ionic Water-Soluble Media with Barium Sulfate1 lodinated water-soluble compounds have been widely recommended as the most suitable contrast media for diagnosis of gastrointestinal perforations. However, the authors present 6 cases in which mu- cosal tears and transmural perforations of the upper gastrointestinal tract were either unrecognizable or inadequately shown during initial evaluation with methylglucamine diatrizoate. Re-exami- nation with barium sulfate demonstrated the precise location and extent of the per- forations. Reasons for the higher diag- nostic yield of barium studies are ex- plained on the basis of experimental and clinical observations. Index terms: Contrast media, comparative studies (GI system, water-soluble examination, 7[0].1232) Esophagus, perforation, 7[1].710 #{149} Gastrointestinal tract, perforation, 7[0].710 Radiology 144: 231-237, July 1982 1 From the Department of Diagnostic Radiology, Evanston Hospital, Northwestern University, Evans- ton, Ill. Presented at the Sixty-sixth Scientific Assem- bly and Annual Meeting of the Radiological Society of North America, Dallas, Texas, Nov. 16-21, 1980. Re- ceived April 14, 1981; revision requested July 14; revi- sion received July 28 and accepted Aug. 12. See also the papers by Wechsler e’f a!. (p. 239) and Dodds et a!. (p. 439) in this issue. sjh P OTENTIAL side effects of the contrast media used in the radiological evaluation of gastrointestinal perforations have been the subject of extensive investigations (1-5). lodinated water-soluble compounds are widely advocated for this purpose due to their relative safety and rapid resorption following extraluminal leakage (2, 6-9). However, several authors have pointed out that 25-50% of esophageal perfo- rations are not seen during esophagography with aqueous contrast agents (10-12). Considerable difficulty in demonstrating extravasation at the site of gastric perforations has also been reported (12-15). Un- fortunately, we know of no comparative studies of diagnostic accuracy of barium sulfate versus iodinated opaque media. We have reviewed 6 cases in which initial evaluation of the upper gastrointestinal tract with methyiglucamine diatrizoate (Gastrografin, Squibb) failed to disclose mucosal tears and perforations that were clearly evident on re-examination with barium sulfate. MATERIAL AND METHODS Five men and 1 woman ranging from 22 to 67 years of age were studied. In addition, we conducted a simple experiment by obtaining radiographs of two polyethylene tubes with an inner diameter of 1.5 mm. One was filled with a 50% suspension of barium sulfate (50% w/w) and the other with undiluted Gastrografin (37% iodine). Both tubes were placed over a 19-cm-thick Plexiglas phantom to simulate the density of abdominal soft tissues. Serial radiographs were then obtained at 90, 100, 110, and 120 kV and 200 mA and phototimed. A 1.2-mm focal spot and Du Pont Cronex-4 film were used. RESULTS In the phantom experiment, subjective evaluation and densito- metric measurements demonstrated remarkably superior visibility of the barium-filled tube at all four exposures (Fig. 1). This suggests that minimal extravasation and a narrow fistulous tract would also be seen better when opacified with barium instead of Gastrografin in the above concentrations. Our findings in the 6 patients are re- ported below. CASE REPORTS CASE I: This 67-year-old man was undergoing upper gastrointestinal en- doscopy for evaluation of a bleeding duodenal ulcer and suspected cancer of the gastroesophageal junction demonstrated on a recent barium study. Considerable difficulty was encountered during insertion of the flexible