EUS to detect evidence of pancreatic disease in patients with persistent or nonspecific dyspepsia

Background: Mild chronic pancreatitis is difficult to diagnose and the diag nosis is therefore not sought routinely in patients with dyspepsia. The aim of our study was to compare the prevalence of endosonographic pancreatic a bnormalities in patients with dyspepsia and control subjects. Methods: The number of endosonographic abnormalities was compared prospecti vely in patients with dyspepsia and control patients. Patients in whom ther e was any suspicion of pancreatic disease were analyzed separately. Results: Between November 1998 and January 1999, 156 patients with dyspepsi a were compared with 27 control patients. The groups were similar except th at control patients were significantly older and more likely to be men. The mean number of endosonographic abnormalities was higher in dyspeptic patie nts than in control patients (mean number of abnormalities 3.3: 95% CI [2.9 , 3.6] vs. 1.9: 95% CI [0.3, 1.7]). The strongest independent predictors of severe endosonographic abnormalities (defined as 5 or more abnormalities) were the presence of suspected pancreatic disease (odds ratio 7.29: 95% CI [2.03, 26.14]) and dyspepsia (odds ratio 7.21: 95% CI [1.99, 26.26]). In th e dyspepsia group, no clinical variables were significant predictors of sev ere abnormalities. However, most patients had nonspecific-type dyspepsia or persistent symptoms after therapeutic trials of acid suppression. Conclusions: Dyspepsia may be an atypical presentation of pancreatic diseas e in patients with persistent or nonspecific symptoms. Endosonography may b e useful to screen for pancreatic disease in patients with persistent dyspe psia.