Gastritis has a wide spectre of definition modalities. Most previous studies have compared symptomatology with histologic gastritis with negative results. We believe that this may be due to inadequate definition criteria and emphasize this point by comparing gastroesophageal reflux with duodenogastric reflux. A prospective randomized trial has been conducted for half a year comparing Sucralfate with a placebo in patients with symptomatological and macroscopical gastritis. Although approximately one hundred patients met the endoscopic criteria, the vast majority could not be included due to well-defined interfering diseases, and thus the material is still too sparse to give any indication of the influence of Sucralfate on endoscopic gastritis, although the preliminary overall results seem promising.