Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Short Barrett's esophagus

Barrett's esophagus is a family of metaplasias: are there genetic markers that correlate with different cell types in Barrett's metaplasia?

H.D. Appelman (Ann Arbor)

Barrett's mucosa would have no interest for us were it not for the fact that it is a precancerous metaplasia; that is, its presence means that a patient with Barrett's has a high risk of developing an adenocarcinoma within that epithelium. There is general agreement that Barrett's mucosa results from chronic gastroesophageal reflux (GER), although the mechanism by which it evolves is not known. Barrett's mucosa has an architecture similar to gastric mucosa. It has superficial and deep compartments (Figure 1). The superficial compartment contains the surface epithelium and the tubular structures that open onto the surface epithelium. The deep compartment has glands that open into the tubules of the superficial compartment. The definition of just what is and is not Barrett's mucosa has gradually evolved, but there is no disagreement that it is a metaplasia and that every epithelial cell within it is metaplastic. In one of the early descriptions of the histologic characteristics of Barrett's mucosa, Paull et al., after examining biopsies from eleven patients, concluded that there were three types of Barrett's mucosa [1]. The first, and the most proximal, was the specialized or distinctive type with goblet cells in the epithelium. The second or junctional type had no goblet cells, was similar to cardiac mucosa with mucus glands and was distal to the first type. The third or fundic type, the rarest and most distal, was a modified body type mucosa with oxyntic glands containing parietal and chief cells but no goblet cells. In no case did the specialized type have fundic glands. Two of the eleven cases had no goblet cells anywhere. For many years, whenever any of these three types of mucosa were found in biopsies of the distal esophagus, they were all considered to be diagnostic of Barrett's mucosa.

In time, it became apparent that in some people, the distal 2 to 3 centimeters of the tubular esophagus are lined by normal proximal gastric mucosa, so in the definition of Barrett's mucosa, this had to be taken into consideration [2]. The new guidelines for the diagnosis of Barrett's mucosa as outlined by a practice committee of the American College of Gastroenterology, state that only mucosa with goblet cells is truly Barrett's mucosa [3]. Therefore, the other two mucosal types of Paull et al., the junctional and fundic types with no goblet cells,

Figure 1. Barrett's mucosa has both superficial and deep compartments

are no longer modifications of Barrett's. Furthermore, based upon these new guidelines, the 2 cases of Paull et al. that had no goblet cells would now not be diagnostic of Barrett's mucosa either. Possibly those non-goblet cell mucosae were really gastric mucosae and came from hiatal hernias. The diagnostic mucosa with goblet cells commonly has deep glands, almost of all of which are cardiac type mucus glands, and it is extraordinarily rare for oxyntic glands to be present (Figure 2). In addition, a third type of gland has been reported to occur in Barrett's mucosa. In this type the constituent cells are identical to pancreatic acinar cells, a phenomenon known as "pancreatic acinar metaplasia". In one study, this was found in biopsies from 8 of 120 patients with Barrett's mucosa, but from the authors' descriptions, these pancreatic acinar cells were not found in the diagnostic, goblet cell containing Barrett's mucosa but in the modified gastric mucosae that are no longer thought to be truly Barrett's mucosae [4]. Therefore, it now appears that the only gland to be found in diagnostic Barrett's mucosa is the mucous gland. As mentioned above, the superficial compartment of Barrett's mucosa contains the surface epithelium and the underlying tubules that have been called "pits" of "crypts" but they really don't have an established name. The superficial compartment is where the goblet cell, the diagnostic marker of Barrett's mucosa, is found, but the goblet cell is not the most common cell here. Goblet cells are scattered among tall columnar cells that superficially resemble the cells of the gastric mucosal surface and those lining gastric pits or foveolae (Figure 3). These have been designated as "intermediate cells." Some of these intermediate cells contain gastric type neutral mucins, butothers contain acid mucins that are clearly non-gastric [5, 6]. The acid

Figure 2. Three kinds of glands have been described as occurring in Barrett's mucosa, but only the cardiac type really is present.

Figure 3. The surface epithelium and that of the superficial tubules has a mixture of cell types including the goblet cells and the columnar cells between them, the intermediate cells.

mucin containing intermediate columnar cells stain light blue with the H and E and intensely blue with the alcian blue stain, and as a result, they have been given the nickname of "columnar blues". The ultrastructural features of the whole family of intermediate cells have been analyzed [7]. As suggested by light microscopic findings, ultrastructurally, some intermediate cells resembled the cells of normal gastric pits and necks. Others had features of both gastric surface mucus cells and intestinal absorptive cells with variable amounts of gastric type mucous granules and intestinal type microvilli. Still other cells had a mixture of gastric type and goblet cell type mucus. In addition, many Barrett's segments have islands of squamous mucosa, some, or even most, of which are metaplastic, thus adding a secondary metaplasia to the Barrett's itself (Figure 4). This squamous metaplasia probably accounts for many of the endoscopic squamous islands and the occasional endoscopic appearance ofshortening of the Barrett's segment. We have no idea about which patients with chronic GER are at risk to develop Barrett's mucosa. Finding Barrett's mucosa is always an accident, because it causes no specific symptoms other than the reflux symptoms it shares with reflux disease in general. As a result, when we see it, it is already an established phenomenon. Furthermore, once it has become established, its extent doesn't seem to lengthen [8]. As a result of these features, we have no clue as to how it evolves. We assume that it develops after the distalesophageal squamous mucosa has been denuded by reflux, but no one has followed this progression by sequential endoscopies and tissue sampling. We do not know the cells that give rise to the Barrett's metaplasia, although there have been severalcandidates, including the proximal gastric mucosa, the basal cells of the squamous mucosa and the cells lining the submucosal gland ducts (Figure 5). A distinctive

Figure 4. Squamous metaplasia often occurs in Barrett's mucosa, both as small foci or islands or as large areas. In both situations, the Barrett's tubules and glands are beneath this metaplastic squamous epithelium

Figure 5. Three proposed precursors for Barrett's mucosa include the cells lining the submucosal glandducts, cardiac mucosa distally and basal cells in thesquamous epithelial cells proximally.

Figure 6. Another proposed precursor is thismultilayered epithelium that has features of both squamous and columnar epithelium. In these photos, the epithelium lies over submucosal gland ducts.

cuboidal epithelium has been described that seems to have features of both squamous and columnar cells, and which had been proposed as the precursor of the Barrett's epithelium [9] (Figure 6). There is no denying the fact that this epithelium can be found in Barrett's, and that it seems to occur at the junction of Barrett's and squamous epithelium. However, it resembles the immature form of squamous metaplasia that is common in the uterine cervix, soperhaps it is not the precursor of the Barrett's, but of the superimposed squamous metaplasia.

This discussion of cell types in Barrett's mucosa is all well and good, but it doesn't answer the question as stated in the title of this discussion, namely whether there are genetic markers that correlate with the individual cell types. Clearly, there are different genetic expressions in the different cells, just as such differences exist in all specialized cells in the body. We know that the columnar cells in Barrett's have different cytokeratin profiles and different enzyme expressions than does the normal squamous epithelium, and these are also different than normal gastric mucosa. We recognize that these differences must be the result of different gene expressions. However, the specific genes are involved in the development of the individual metaplastic cell types and which genetic changes are involved in the initiation of the first Barrett's metaplasia are completely unknown. Obviously, we know a great deal about the different cells in Barrett's mucosa, but we know nothing about how they got there.


1. Paull A, Trier JS, Dalton MD, et al. This histologic spectrum of Barrett's esophagus. N Engl J Med 1976;295:476-480.

2. Haggitt RC. Barrett's esophagus, dysplasia, and adenocarcinoma. Hum Pathol 1994;25:982-993.

3. Sampliner RE and the Practice Parameters Committee of the American College of Gastroenterology. Practice guidelines on the diagnosis, surveillance, and therapy of Barrett's esophagus. Am J Gastroenterol 1998;93:1028-1032.

4. Krishnamurthy S, Dayal Y. Pancreatic metaplasia in Barrett's esophagus. An immunohistochemical study. Am J Surg Pathol 1995;19:1172-1180.

5. Chaves P, Cardoso P, de Almeida JCM, et al. Non-goblet cell population of Barrett's esophagus: an immunohistochemical demonstration of intestinal differentiation. Hum Pathol 1999;30:1291-1295.

6. Offner FA, Lewin KJ, Weinstein WM. Metaplastic columnar cells in Barrett's esophagus: a common and neglected cell type. Hum Pathol 1996;27:885-889.

7. Levine DS, Rubin CE, Reid BJ, Haggitt RC. Specialized metaplastic columnar epithelium in Barrett's esophagus. A comparative transmission electron microscopic study. Lab Invest 1989;60:418-432.

8. Cameron AJ, Lomboy CT. Barrett's esophagus: age, prevalence, and extent of columnar epithelium. Gastroenterology 1992;103:1241-1245.

9. Boch JA, Shields HM, Antonioli DA, et al. Distribution of cytokeratin markers in Barrett's specialized columnar epithelium. Gastroenterology 1997;112:760-768.

Publication date: August 2003 OESO©2015