Endoscopic Ultrasonography for the Diagnosis of Chronic Pancreatitis

Background: Previous studies have shown the Endoscopic Ulstrasonongraphy (EUS) is very sensitive for the detection of morphologic abnormailities due to chronic pancreatitis. Unfortunately there are uncertainities about the specificity of EUS findings, particularly in the early stages of chronic pancreatitis. Since the “gold standard” for the diagnosis of chronic pancreatitis (tissue diagnosis) is lacking, one way to validate findings is to do clinical follow up. Aim: Our aim is to understand if in a clinical context suspect for chronic pancreatitis, patients who have a normal ERCP but an EUS that shows features of chronic pancreatitis, will develop imagilogic or clinical characteristic features of chronic pancreatitis, comfirming a higher sensitivity of EUS, or on the other hand, if they have false positive EUS for chronic pancreatitis. Method: In a retrospective review of the EUS database from a 4-year period (January of 1996 - Dec 1999), we found 240 patients less than 65 years old and clinical symptoms suggestive of chronic pancreatitis that had both EUS and ERCP. Patients with malignancy were excluded. Of these 240 patients, 48 (20%) had discrepancy between EUS and ERCP findings, withy normal ERCP but EUS features of moderate/high probability for chronic pancreatitis (>3 features on EUS). Chart review or contact by phone was used to document pancreatitis. Patients were asked if they were diagnosed with pancreatitis. If they responded yes, two questions were asked verbatim without coaxing. These questions included when the diagnosis was made and how. Results: 48 Patients (Female 34; mean age 42 (range 19 - 62 years) were reviewed. The average follow-up time was 8.4 years (range 7 - 9.6 years). 29/48 (60%) patients were contacted or chart reviewed, 19 were lost to follow-up. Of those suspected of having early chronic pancreatitis in which EUS was the only abnormality, 55% progressed clincially such that the diagnosis could be established. 13/29 (45%) of the patients did not have a diagnosis of pancreatitis on follow-up. Conclusion: EUS can be used in the preliminary diagnosis of pancreatitis. Before 2001 years ago, we use 3 or more criteria on EUS to diagnose chronic pancreatitis. We decided based on histology and the fact that we would rather have a higher specificity rather than sensitivity to label patients as “chronic pancreatitis”, we changed our criteria from 3 or more to 5 or more. Thus, some of the 45% that did not evolve into clear chronic pancreatitis may have been read as normal by todays criteria.