In theory, the diagnosis of Barretts esophagus should be fairly simple; namely the presence of intestinal metaplasia anywhere in the tubular esophagus. The evolving concept of short segment Barretts esophagus makes the diagnosis of Barrett’s esophagus a bit more problematic. The difficulty in diagnosing short segment Barrett’s esophagus is because the precise junction of the stomach and the esophagus may be difficult to determine endoscopically due to the presence of a hiatal hernia, inflammation, and disassociation between the squamocolumnar junction and the lower esophageal sphincter zone in normal individuals. As such, an irregular Z-line may appear abnormal to one observer and normal to another one. Biopsies from a normal gastroesophageal junction may also reveal intestinal metaplasia which is presumed to represent intestinal metaplasia of the cardia. These differences are not merely semantics. The dysplasia risk in short segment Barrett’s esophagus is significantly greater than in intestinal metaplasia of the cardia. As such, surveillance is recommended for patients with short segment Barrett’s esophagus but not for patients with intestinal metaplasia of the cardia. Despite the widespread emphasis on the importance of specialized columnar epithelium as the sine qua none of Barrett’s esophagus and increased cancer risk, many pathologists continue to classify gastric metaplasia without intestinal metaplasia as Barrett’s esophagus thereby increasing patient alarm and needlessly subjecting many individuals to unnecessary surveillance endoscopy.
Techniques such as methylene blue staining may be helpful in directing biopsies, especially in patients with suspected short-segment Barretts esophagus, since the stain is typically picked up by specialized columnar epithelium. Another possible option is to examine the immunohistochemical cytokeratin staining patterns of biopsies obtained from questionable areas of Barrett’s esophagus. Recent work suggests that characteristic staining patterns may reliably distinguish between intestinal metaplasia of the stomach and the esophagus.
Therefore, at the time of endoscopy, landmarks should be carefully defined: the diaphragmatic hiatus, esophagogastric junction, and squamocolumnar junction. If the squamocolumnar junction is above the level of the esophagogastric junction, biopsies should be obtained. If intestinal metaplasia is present, defined by goblet cells seen with combined hematoxylin and eosin-alcian blue pH 2.5 stains, the patient is considered to have Barretts esophagus and should be placed in a surveillance program. Biopsies of the squamocolumnar junction should not be routinely obtained in clinical practice if it is at the level of the gastroesophageal junction. Goblet cells found at this level should be considered to be diagnostic of intestinal metaplasia of the gastric cardia, a condition with an unclear cancer risk and one in which cancer surveillance is not yet recommended. These patients may well be infected with H. pylori, a diagnosis that can be confirmed with biopsies of the more distal antrum and fundus or noninvasive techniques such as the urea breath test, stool antigen studies or serology.
