Clinical value of endoscopy and histology in the diagnosis of duodenogastric reflux disease.

Abstract

The endoscopic observation of a bile lake in the stomach, antral gastritis, or ulcerations and the histologic finding of foveolar hyperplasia or chronic gastritis have been implicated as indicators of excessive duodenogastric reflux. The accuracy of these criteria was evaluated in 135 patients with nonspecific symptoms in the foregut suggestive of duodenogastric reflux and no evidence for alcohol- or drug-induced gastric mucosal injury. The presence of excessive duodenogastric reflux was objectively determined by means of both gastric pH monitoring and cholescintigraphy with cholecystokinin stimulation. Endoscopy showed antral gastritis in 67 patients, gastric ulcers in 19, and a bile lake in the stomach in 39 (total of 135 patients). Of 90 patients who underwent biopsy, histologic findings showed foveolar hyperplasia in 26, chronic gastritis in 19, and active gastritis in 28 patients. The latter condition was associated with Helicobacter pylori in 20 patients. When gastric pH monitoring, cholescintigraphy, or both were used as "gold standard," the sensitivity, specificity, accuracy, and positive predictive value of endoscopic and histologic criteria to diagnose the presence of excessive duodenogastric reflux were poor except in the rare case of active gastritis but no Helicobacter pylori. The presence of duodenogastric reflux disease cannot be accurately diagnosed with endoscopic or histologic criteria. The diagnosis should be made with objective techniques, particularly when surgical therapy is considered.