Reply

Duodenal ulceration and postoperative recurrence SIR,-We have read with interest the leading article by Mr R M Kirk (Gut 1988; 29: 1625-7). He draws attention to reports of the presence of parietal cells in the duodenum and suggests that these may enable 'susceptible mucosa to come into contact with freshly secreted acid that has not yet been partially diluted, adsorbed, absorbed, and neutralised'. During the past six years we have had a particular interest in gastric metaplasia in association with duodenal ulceration and have devised scoring systems for recording light and electron microscopical changes. 'During this time several hundred duodenal biopsies have been examined and we have not yet seen any oxyntic cells in the presence of gastric metaplasia. Typically the mucosal surface is flat, without any gastric pits and is covered by epithelial cells secreting PAS staining mucus. This stresses the importance of understanding the exact meaning of any references to the presence of gastric mucosa in the duodenum. It is important to distinguish between gastric metaplasia and the presence of heterotopic gastric mucosa. The former often involves widespread areas of mucosa covered with mucus secreting surface epithelial cells staining with PAS accompanied by varying degrees of inflammatory cell infiltration and frequently the presence of Campylobacter pylori. The lattef involves scattered small islands of parietal cells, usually in association with Brunner's glands lying superficial to the muscularis mucosae, sometimes with occasional chief cells or small clusters of surface epithelial cells. Both are reported as occurring more frequently in association with duodenal ulceration. Gastric metaplasia is thought to be either a defence mechanism or a manifestation of mucosal damage in response to hyperacidity, whereas the presence of heterotopic mucosa may be developmental and a possible course of localised hyperacidity contributing to ulcer formation.