Abstract
Assessments of the diagnostic value of 4 tests in 22 patients with a proven carcinoma of the pancreas were made. ERCP, US, CT and RI were interpreted as correct, suspicious and incorrect according to the criteria described previously by us and others. ERCP and CT were most sensitive in diagnosing pancreatic cancer (77.3%, 72.2%, respectively). These 4 tests offered complementary, correct or suspicious information in the evaluation of pancreatic cancer. An indirect sign on US and CT in patients with jaundice is dilatation of the intrahepatic biliary tree. All the 4 tests are useful for the diagnosis in patients with or without jaundice. Most of tumors are not found at a resectable stage. In this series only 4 patients had a resectable pancreatic cancer. ERCP gave the correct diagnosis in these 4 patients. ERCP is the best means of detecting a resectable tumor. Most patients, however, ex-perienced vague, nonspecific symptoms at an early stage in the disease process. If an effec-tive diagnostic procedure were used at an earlier stage, not only, would the prognosis after resection be improved, but also a further group of patients who would otherwise progress to non-resectable disease could be identified and offered radical surgical treatment. We proposed a decision tree for the diagnosis of pancreatic cancer. A more sensitive diagnostic modality for the correct diagnosis of pancreatic cancer remains to be established. CT combined with angiography, RI with subtraction and direct magnification during ERCP are under investigation.
