Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Diagnosis

What is the definition of the junction between the esophagus and stomach from a histologic perspective?

P. Chandrasoma (Los Angeles)

Defining the junction between the esophagus and stomach seems easy. The esophagus is a tubal structure and the stomach is a pouch; surely, it is not difficult to define where the tube ends and the pouch begins. However, if the lower part of the tube dilates, the gastroesophageal junction (GEJ) becomes less definite. Because Barrett's esophagus is presently defined as "a change in the esophageal epithelium of any length that can be recognized at endoscopy and is confirmed to have intestinal metaplasia by biopsy" [1], it is critically important to have an accurate definition of the GEJ. Without an accurate definition of the GEJ that is not subject to interobserver variation, it is impossible to distinguish esophageal from gastric epithelium in the junctional region.

The GEJ has been defined by endoscopic and gross pathologic criteria. The landmarks that have been used to define the GEJ historically have been the squamocolumnar junction, the distal limit of the tubal structure, the diaphragmatic crus, the distal limit of the lower esophageal sphincter (LES), and the proximal limit of the gastric rugal folds. Of these, the only ones presently used are the distal limit of the LES and the proximal limit of the gastric rugal folds, with the latter criterion achieving the status of currently accepted standard in the past few years.

The only histologic definition of the GEJ was suggested by Norman Barrett in 1950 [2], who defined the esophagus as that part of the foregut that is lined by squamous epithelium. The glandular epithelium lined tubal structure distal to the squamocolumnar junction present in some patients became a tubular stomach by this definition. In 1957, Barrett reversed himself and recognized the existence of the "columnar lined esophagus" that became the original basis for Barrett's esophagus [3]. With this reversal, the only histologic definition of the GEJ that has ever been proposed disappeared.

The purpose of this paper is to define the GEJ histologically.

Accurate definition of the gastroesophageal junction is a problem

The distal esophagus passes from the thorax into the abdomen at the diaphragmatic hiatus. At this point, the intraluminal pressure in the esophagus changes from a mean negative intrathoracic pressure to a positive intra-abdominal pressure. This pressure change is not demonstrable, however, because the distal 5 cm of the esophagus is the LES, whose high resting pressure masks the intraluminal pressure. Despite the fact that the LES overwhelms intraluminal pressure in the distal esophagus, it is important to recognize that the intraabdominal distal esophagus below the diaphragm has a potential positive intraluminal pressure.

Reflux causes permanent damage to the LES, initially manifested as a loss of intraabdominal length followed by a loss of mean pressure and overall length [4]. Shortening of the LES in reflux makes the distal limit of the LES a variable point and therefore unreliable as a criterion to define the GEJ. Loss of resting high pressure in the intra-abdominal part of the LES has the effect of causing dilatation of this part of the distal esophagus because of its positive intraluminal pressure ("gastricization of the esophagus"), making its anatomic definition from the gastric pouch more difficult. Conversion of the distal esophagus from a tube to a dilated pouch-like structure makes the end of the tubal esophagus an unreliable criterion for defining the GEJ.

Gastric rugal folds are characteristic of the oxyntic mucosa of the proximal stomach. When glandular metaplasia of the esophageal squamous epithelium occurs, the metaplastic glandular mucosa is typically flat and lacks rugal folds. The proximal limit of the rugal folds therefore remains as a permanent marker of the GEJ. However, this criterion is reliable only if the metaplastic mucosa does not develop rugal folds. This has not been proven. As such, the proximal limit of the rugal folds represents the best available evidence of the GEJ rather than absolute proof of the location of the GEJ.

Normal histology of the gastroesophageal junction

The prevailing view of normal histology is that there is a 2-3 cm length of cardiac (also termed junctional) mucosa between the oxyntic mucosa of the stomach and the squamous epithelium of the esophagus [5]. This view has been shown to be incorrect by two recent autopsy studies. In our study [6], we found that 56% of patients did not have any cardiac mucosa even when the entire circumference of the squamocolumnar junction was examined histologically. Pure cardiac mucosa is characterized by glands containing mucous cells only (Table I). All patients had oxynto-cardiac mucosa, characterized by glands containing a mixture of parietal and mucous cells. However, oxynto-cardiac mucosa was not present in the entire circumference of the squamocolumnar junction in 50% of patients. The combined cardiac mucosa + oxynto-cardiac mucosa length was < 0.5 cm in 76% of patients. Kilgore et al. [7], in an autopsy study of children reported that the mean length of cardiac mucosa was 0.18 cm with a range of 0.1-0.4 cm. These data indicate that cardiac mucosa is either absent or extremely small in extent at the squamocolumnar junction. Data has also been presented that the length of cardiac mucosa present at the squamocolumnar junction increases with the age of the patient [6, 8] and with increasing amounts of reflux [9].

Table I. Criteria for classifying epithelial types at the gastroesophageal junctional region.

For practical purposes, therefore, normal histology of the squamocolumnar junction invariably consists of only pure oxyntic mucosa and squamous epithelium in all parts of the circumference. The presence and extent of cardiac mucosa and oxynto-cardiac mucosa at the squamocolumnar junction is variable, the cardiac mucosa + oxynto-cardiac mucosa length is < 0.5 cm in the majority of normal people, and the cardiac mucosa length increases with age and acid exposure. Evidence will be presented in my answer to the next question that cardiac mucosa and oxynto-cardiac mucosa are abnormal glandular mucosae at the squamocolumnar junction rather than normal structures.

Histologic definition of the gastroesophageal junction

If it is accepted that the only normal mucosal types at the squamocolumnar junction are pure oxyntic mucosa and squamous epithelium, the definition of the GEJ and the pathologic changes of reflux become easy to understand. Normality is a stomach that is lined by oxyntic mucosa which transitions to an esophagus that is lined by squamous epithelium at the GEJ. Reflux damages the squamous epithelium of the esophagus and the LES resulting in the interposition of a reflux damaged distal esophagus between the stomach and the undamaged esophagus. This reflux-damaged distal esophagus is most accurately recognized histologically by the presence of cardiac mucosa and oxynto-cardiac mucosa that is interposed between the esophageal squamous epithelium and gastric oxyntic mucosa. Reflux-induced cardiac mucosa and oxynto-cardiac mucosa must be derived from squamous epithelium of the distal esophagus because gastric oxyntic mucosa is not damaged by reflux. The proximal limit of histologically defined pure oxyntic mucosa remains unaltered and represents a permanent marker for the original GEJ. A histologic definition of this region can now be proposed: the GEJ is the proximal limit of pure oxyntic mucosa; everything proximal is esophagus.

When biopsies are taken from the "proximal limit of the rugal folds" it is common to see two epithelial types: oxyntic mucosa and oxynto-cardiac mucosa. When oxynto-cardiac mucosa contains numerous parietal cells, it can resemble oxyntic mucosa, and it is very likely that is can produce folding that can mimic the gastric rugal folds. Histology is much more accurate than endoscopy in differentiating oxyntic mucosa from oxynto-cardiac mucosa, and is therefore superior as a criterion to define the GEJ.


1. Sampliner RE, 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.

2. Barrett NR. Chronic peptic ulcer of the oesophagus and "oesophagitis". Br J Surg 1950;38:175-182.

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

4. Zaninotto G, DeMeester TR, Schwizer W, Johansson KE, Cheng SC. The lower esophageal sphincter in health and disease. Am J Surg 1988;155:104-111.

5. Hayward J. The lower end of the esophagus. Thorax 1961;16:36-41.

6. Chandrasoma PT, Der R, Ma Y, Dalton P, Taira M. Histology of the gastroesophageal junction;an autopsy study. Am J Surg Pathol 2000;24:402-409.

7. Kilgore SP, Ormsby AH, Gramlich TL, Rice TW, Richter JE, Falk GW, Goldblum JR. The gastric cardia: fact or fiction? Am J Gastroenterol 2000;95:921-924.

8. Ormsby AH, Goldblum JR, Kilgore SP, Richter JE, Rice TW, Falk GW, Gramlich TL. The frequency and nature of cardiac mucosa and intestinal metaplasia (IM) of the esophagogastric junction (EGJ): a population based study of 223 consecutive autopsies. Gastroenterology 1999;116:A273.

9. Chandrasoma PT, Lokuhetty DM, DeMeester TR, Bremner CG, Peters JH, Oberg S, Groshen S. Definition of histopathologic changes in gastroesophageal reflux disease. Am J Surg Pathol 2000;24:344-351.

Publication date: August 2003 OESO©2011