Hydrospiral CT of the pancreas in thin section technique

PURPOSE: To evaluate a new technique for imaging the pancreas (spiral hydro-CT) based on a combination of pharmacological intestinal paralysis and water distension of the stomach and duodenum with specific reference to tumor detection rate, differentiation of malignant versus benign tumors, differential diagnosis and assessment of tumor resectability. MATERIAL AND METHODS: In an open prospective study, 151 patients with a suspected pancreatic neoplasm based on clinical, laboratory or other imaging data were examined between May 94 and October 95. Our newly developed Hydro-CT methodology included intravenous injection of 40 mg N-butylscopolaminium bromide (Buscopan) for intestinal paralysis, gastric and duodenal wall distension by oral administration of an average of 1.51 warm tap water, 30 degrees RAO patient positioning, an individualized contrast injection technique as determined beforehand by time-to-peak measurement in the portal vein and thin-slice spiral CT (3 mm increment, 6 mm table feed and 3 mm secondary reconstruction). A detailed evaluation form was used to assess (1) tumor detection rate, (2) differentiation of malignant versus benign disease, (3) differential diagnosis, and (4) accuracy of assessment of resectability by identification of infiltration into adjacent organs and vessel structures relevant for resectability such as splenic, superior mesenteric, portal vein and celiac trunk, superior mesenteric, splenic and hepatic arteries. As the gold standard for positive tumor detection surgery and microscopic diagnosis were used, and for negative tumor detection an event-free survival of 6 months. RESULTS: Almost all examinations were well tolerated. In only 4% was on-site administration of a gastric tube required because of vomiting. In 2% of the patients a slight allergic reaction to the contrast medium was seen. The prevalence of a pancreatic neoplasm was 38%. In tumor detection Hydro-CT reached an overall accuracy of 97.4% with a sensitivity of 100% and a specificity of 95.9%. In the differentiation of benign versus malignant disease Hydro-CT reached an overall accuracy of 89.7% with a sensitivity of 92.5% and a specificity of 83.3%. The prevalence of a pancreatic carcinoma was 24%; 4% other malignant tumors were found (distal common bile duct carcinoma, cystadenocarcinoma). Fifty-eight patients underwent surgical exploration. In those assessment of resectability reached an overall accuracy of 95% with a sensitivity of 90.5% and specificity of 100%. CONCLUSION: The new technique of Hydro-CT based on thin slice and spiral methodology, including pharmacological intestinal paralysis and water distension, results in a high tumor detection rate and increases sensitivity and specificity of tumor differential diagnosis and of assessment of resectability.