An endoscopic pancreatic function test with cholecystokinin-octapeptide for the diagnosis of chronic pancreatitis.

BACKGROUND & AIMS: Current pancreatic function tests are cumbersome and unavailable to the clinical gastroenterologist. We have developed a function test that can be modified to a purely endoscopic collection method (ePFT). The aim of this study was to compare the endoscopic and traditional Dreiling tube collection methods.
METHODS: Two separate groups of healthy subjects and patients with chronic pancreatitis underwent pancreatic function testing. One group underwent the endoscopic collection method (ePFT). Intravenous cholecystokinin (CCK 40 ng x kg(-1) x h(-1)) was started in preprocedure area. Duodenal fluid was collected with upper endoscope during endoscopy at 30, 40, 50, and 60 minutes during infusion. Another group underwent the traditional Dreiling collection method. Intravenous CCK was started in postprocedure area after endoscopic tube placement. Duodenal fluid was collected at 0, 20, 40, 60, and 80 minutes during infusion. Lipase concentration was determined (IU/L) on laboratory autoanalyzer.
RESULTS: Seventy-three patients were studied. Thirty-four underwent endoscopic collection and 39 underwent Dreiling collection. The mean peak lipase concentrations (+/- standard deviation) for healthy subjects and patients with chronic pancreatitis in the endoscopic collection method group were 1612500 +/- 556152 IU/L and 369594 +/- 281624 IU/L, respectively (P < 0.001). The mean peak lipase concentrations (+/- standard deviation) for healthy subjects and patients with chronic pancreatitis in the Dreiling tube collection method group were 1670324 +/- 786731 IU/L and 478956 +/- 406061 IU/L, respectively (P < 0.001). There was no statistical difference between collection methods at distinguishing healthy subjects and patients with chronic pancreatitis. Receiver operating characteristic curves (ROC) for the endoscopic and Dreiling collection methods were 0.993 (standard error of mean, 0.009) and 0.921 (standard error of mean, 0.041). A lipase concentration cut point of 810600 IU/L distinguishes healthy subjects from patients with chronic pancreatitis with a sensitivity and specificity of 92% and 95%, respectively. The ePFT was safe, short in duration, minimized costs (US dollars 1890 vs. US dollars 2659), required small amounts of fluid for analysis (1-3 mL), and eliminated radiation exposure.
CONCLUSIONS: Analysis of timed endoscopic aspirations of pancreatic juice after hormonal stimulation can distinguish healthy subjects from patients with chronic pancreatitis. This new endoscopic collection method (ePFT) is less cumbersome and more time efficient, when compared to traditional collection methods. The ePFT broadens the availability of function testing to the practicing clinical gastroenterologist.