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

How to classify patients with endoscopic Barrett's esophagus but absent intestinal metaplasia on biopsy?

G. Chejfec, S. Sontag, T. Schnell (Hines)

The definition of Barrett's esophagus has been in a constant state of change since its original description by Norman Barrett, who defined it as "the lower esophagus lined by columnar epithelium" [1]. The type of columnar epithelium was described 20 years later in a study of 11 patients with Barrett's esophagus, in whom three types of epithelia were identified: gastric fundic-type, junctional-type, and specialized intestinal-type with goblet cells [2]. Currently, the diagnosis of Barrett's esophagus is established only with the presence of specialized intestinal epithelium, e.g. goblet cells [3]. Recently, the definition of Barrett's esophagus that is based solely on the presence of goblet cells has been challenged [4]. In a detailed study of 9 patients who underwent esophagectomy for Barrett's-associated adenocarcinoma or high-grade dysplasia, more than 80% of all sections examined contained not only goblet cells, but columnar cells that were alcian-blue pH 2.5 positive. Columnar cells that are alcian-blue positive confirm the presence of intestinal-type sialomucins, which in turn support the concept of intestinal metaplasia (IM) in the absence of goblet cell morphology. Because of similarities between alcian-blue positive columnar cells and the cells found in gastric incomplete IM, the authors suggested that goblet cell-negative metaplasia also be considered as a risk lesion for adenocarcinoma [4]. Goblet-negative columnar cells identified in Barrett's esophagus have also been studied by other investigators, who used immunohistochemistry to determine the presence of phenotypic markers of intestinal differentiation [4]. The two markers selected were sucraseisomaltase and dipeptidylpeptidase, both of which are proteins expressed by absorptive cells of normal adult and fetal intestinal epithelium, both of which are present in the apical portion of the cell (with dipeptidylpeptidase also present in the cytoplasm), and both of which are absent in normal gastric and esophageal epithelium. Twelve surgical specimens of Barrett's mucosa containing adenocarcinoma or high-grade dysplasia were examined using monoclonal antibodies against both proteins (antigens). The results clearly demonstrated that both antigens were present in the malignant as well as the adjacent columnar cells, but were not present in the goblet cells. The authors concluded that nongoblet columnar cells expressing intestinal antigens represent an incomplete form of IM that, by extension, may have the potential for malignant transformation of Barrett's esophagus [5]. These results confirmed the findings by the group that six years earlier was the first to demonstrate expression of the sucrase-isomaltase gene in the majority of malignant and columnar cells [6].

In view of the ever increasing body of literature regarding the occurrence of gobletnegative columnar cells along with goblet-positive adenocarcinoma, we propose a new classification of Barrett's esophagus comprising two distinct groups: goblet cell-positive IM, and goblet cell-negative IM. The first group (goblet cell-positive IM) would include patients with pure goblet cell metaplasia as well as those with the combination of goblet cell-positive and goblet cell-negative metaplasia. The second group (goblet cell-negative IM) would be limited to those patients who have definite columnar cells that are clearly in the tubular esophagus and that are truly goblet cell negative.

Such goblet cell-negative columnar cells have received increasing attention in recent years. They are thought to be found:
- mostly in short segments at the esophagogastric junction,
- in predominantly white populations,
- in older patients,
- in patients without necessarily a history of gastroesophageal reflux disease.

Thus far, an association with dysplasia or carcinoma has not been a major issue [7, 8].

We propose that it is precisely the group of patients with goblet cell-negative IM confined to the esophagogastric junction (EGJ) that is responsible for the subset of EGJ adenocarcinomas that does not fit neatly into the current definition of Barrett's esophagus.


1. Barrett NR. The lower esophagus lined by columnar epithelium. Surgery 1957;41:881-894.

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

3. Spechler SJ, Goyal RK. The columnar-lined esophagus, intestinal metaplasia, and Norman Barrett. Gastroenterology 1996;110:614-621.

4. Offner FA, Weinstein WM, Lewin KJ. Columnar, non-goblet cell metaplasia: a neglected feature in the diagnosis of Barrett's esophagus. Gastroenterology 1995;108:A180.

5. Chaves P, Cardoso P, Crespo Mendes de Almeida J, Dias Pereira A, Nobre Leitao C, Soares J. Non-goblet cell population of Barrett's esophagus; an immunohistochemical demonstration of intestinal differentiation. Hum Pathol 1999;30:1291-1295.

6. Wu GD, Beer DG, Moore JH, Orringer MB, Appelman HD, Traber PG. Sucrase-isomaltase gene expression in Barrett's esophagus and adenocarcinoma. Gastroenterology 1993;105:837-844.

7. Trudgill NJ, Suvarna SK, Kapur KC, Riley SA. Intestinal metaplasia at the squamo-columnar junction in patients attending for diagnostic gastroscopy. Gut 1997;41:585-589.

8. Spechler SJ, Zeroogian JM, Antonioli DA Wang HH, Goyal RK. Prevalence of metaplasia at the gastro-esophageal junction. Lancet;344:1533-1536.

Publication date: August 2003 OESO©2015