[Diagnosis of chronic pancreatitis. Studies of duodenal juice after stimulation with the secretin-ceruletide test. Decision limits and evaluation of various parameters].

187 patients were checked up over 4 years by the secretin-ceruletide test. Independently of the test results they were assigned to various disease groups on the basis of clinical assessment. 131 subjects were divided in a pilot investigation into: subjects with a healthy pancreas (n = 55); subjects with chronic pancreatitis (n = 50); subjects whose pancreatic condition could not be classified clearly (n = 26). 8 parameters were compared by univariate and multivariate statistical procedures in order to confirm or rule out the presence of chronic pancreatitis. The discriminatory power of the following parameters in duodenal fluid proved to be sufficiently high, with less than 15% frequency of misclassification: chymotrypsin (activity) and/or; lipase (activity) and/or; amylase (activity); viscosity. Under routine conditions measurement of the activity of two of these enzymes is sufficient. Their contribution to discrimination proved to be approximately equal. The diagnostic sensitivity and specificity of the parameters bicarbonate, lipase (concentration), trypsin (activity) and volume of duodenal fluid are lower. The classification rules derived from the above pilot group were confirmed by a diagnostic study under routine condition in a test group of 38 patients. Limitation to examining only volume and a maximum of 3 parameters which proved best in distinguishing between patients with chronic pancreatitis and healthy subjects, together with the omission of the first-hour samples after a secretin bolus, considerably reduced laboratory workload without altering the discriminatory power of the secretin-ceruletide test.