—  SHORT COURSE #03  —

Gastrointestinal Pathology: Controversies in Non-neoplastic Biopsy Interpretation
Moderators: Robert D. Odze, Marie E. Robert, Donald A. Antonioli

Section 1 - Diagnostic Problems in Biopsy Interpretation of the Gastroesophageal Junction

Robert D. Odze
Chief, Gastrointestinal Pathology Service, Associate Professor of Pathology
Brigham and Women's Hospital, Harvard Medical School, Boston, MA


Introduction
The gastroesophageal junction (GEJ) is a poorly defined anatomic area that represents the junction between the distal esophagus and the proximal stomach (cardia). The GEJ is an area that is commonly exposed to the injurious effects of gastroesophageal reflux disease (GERD) and H.pylori infection [1, 2]. These are the two most common pathologic disorders affecting this region. In fact, both of these disorders may result in chronic inflammation and the subsequent of development of intestinal metaplasia (IM), which increases the risk of neoplasia [3]. Over the last several decades, there has been a dramatic rise in the incidence of adenocarcinoma of the GEJ region, which is presumed to be related to GERD, H.pylori or both [4].

This review will focus on the clinical, pathologic, and pathogenetic aspects of GERD and H.pylori-induced inflammation of the GEJ region. For the purpose of this review, the GEJ is considered the mucosal area involving the distal 1 cm of esophagus and the proximal 1 cm of stomach. Unfortunately, there are many limitations in performing studies on diseases of the anatomic region [1, 5]. This is, in part, related to problems in evaluating anatomic landmarks of the GEJ endoscopically, inadequate standardization of endoscopic and pathologic methods, and inconsistent use of definitions and terminology used to define the histology and pathology of this region. For instance, there is a lack of consistence in the literature with regard to patient populations studied, methods of endoscopy utilized, sites of tissue sampling, and the use of adjunctive tests in defining IM in mucosal biopsies. Nevertheless, the information in this review article represents an evidence-based analysis, and provides information that pathologists may use to differentiate GERD from H.pylori infection, and determine the cause of IM, in the GEJ region.

Histology of the Gastroesophageal Junction
The esophagus is lined by non-keratinizing stratified squamous epithelium, and contains scattered mucous glands within the lamina propria similar to those that reside in the gastric cardia [6]. In addition, the esophagus contains multiple linear rows of salivary gland-like mixed mucous/serous glands in the submucosa which empty their contents into the lumen by a duct system lined partly by cuboidal, and partly by squamous, epithelium. A mild periductal infiltrate of mononuclear cells is commonly present surrounding the submucosal glands and ducts. The stomach is lined by mucinous columnar epithelium and contains either pure oxyntic glands, or pure mucous glands, in the deep portion of the mucosa of the corpus and antrum/pylorus, respectively. A mixture of oxyntic and mucous glands is often present at the junction between these two gastric compartments (Figure 1). The true anatomic GEJ corresponds to the most proximal aspect of the gastric folds, which represents an endoscopically apparent transition point in the most individuals [7]. Unfortunately, this anatomic landmark is often obscured in patients with hiatus hernia, which is a major cause of GERD. In "normal" individuals, the anatomic GEJ also corresponds to the histologic transition point between the esophageal squamous epithelium and the gastric mucinous columnar epithelium. This transition point is termed the "Z"-line [6]. However, many, if not most, adults, particularly those with either physiologic or pathologic GERD, have a proximally displaced "irregular" Z-line indicating that the histologic squamocolumnar junction (SCJ) is located above the anatomic GEJ. The histologic characteristics of the short segments of columnar mucosa located above the anatomic GEJ in these individuals are similar to the gastric cardia (see below), being composed of either pure mucous glands or mixed mucous/oxyntic glands, and leads to difficulty in distinguishing columnar metaplasia of the distal esophagus from the true gastric cardia in biopsies from the GEJ region [8, 9]. As discussed further below, this distinction is clinically important since columnar metaplasia of the esophagus is caused by GERD and represents a non-intestinalized type of Barrett's esophagus (BE) and a precursor for the intestinal type of BE, the latter of which has recently been defined by the presence of goblet cells [5, 10].

In contrast, the true gastric cardia is defined as the area of mucosa located distal to the anatomic GEJ and proximal to the portion of stomach composed entirely of oxyntic glands (corpus). Inflammation of this anatomic region is termed "carditis", and is most often caused by H.pylori infection although there is accumulating evidence to suggest that GERD may cause carditis as well [1, 3, 11, 12]. Unfortunately, endoscopists often mistakenly use the term "cardia" to indicate the location of a mucosal biopsy from the GEJ region, even though, in some cases, the biopsy may have actually been obtained from the distal esophagus [3, 8]. Thus, it is often incumbent upon the pathologist to determine the precise location of a particular biopsy of the GEJ region, particularly if it contains IM, which is a major risk factor for the subsequent development of dysplasia and carcinoma. Thus, IM in the GEJ region may represent several diagnostic possibilities; ultrashort segment BE (which is defined as columnar metaplasia of distal esophagus, with goblet cells, of <1.0 cm in length) or chronic carditis with IM. These two conditions have a different etiology, natural history and treatment and are discussed further below [2, 13, 14].

There is controversy regarding the origin and histologic features of the true gastric cardia [1, 11, 15, 16, 17]. Some authors believe that the cardia is normally composed, at birth, of surface mucinous columnar epithelium and underlying oxyntic glands identical to the gastric corpus, whereas other maintain that the true anatomic cardia is normally composed of mucinous columnar epithelium with underlying mucous glands, or mixed mucous and oxyntic glands (mucous/oxyntic glands). Proponents of the former theory believe that the mere presence of mucous, or mixed mucous/oxyntic, glands in the mucosa from the GEJ region indicates that it is metaplastic in origin and developed as a result of GERD [16, 18]. Nevertheless, the results of several studies in neonates, infants and young pediatric patients, all patient groups that have a low incidence of GERD and H.pylori infection, have shown that the true gastric cardia is composed of either pure mucous, or mixed mucous/oxyntic, glands underlying mucinous columnar epithelium in most, if not all, individuals [11, 15, 17, 19, 20]. The area of mucosa distal to the GEJ occupied by this type of epithelium ranges from 1-4 mm in length [15]. For instance, in a study of 30 consecutive pediatric autopsies by Kilgore et al, mucous glands were present on the gastric side of the anatomic GEJ in 100% of cases15. The mean length of mucosa composed of mucous glands was 1.8 mm. Similar findings were reported by Glickman et al in a biopsy study of 74 pediatric patients in which pure mucous glands were identified in the true gastric cardia in 81%, and mixed mucous/oxyntic glands in the remaining 19% of patients [17]. Interestingly, a very small portion of patients show only pure oxyntic glands in the "cardia" region [16, 21]. However, this finding is often focal and does not usually involve the entire circumference of the true gastric cardia.

There is also evidence to suggest that the length of mucosa composed of either pure mucous glands, or mixed mucous/oxyntic glands, increases with age, and is presumed to be related to ongoing "physiologic" GERD [15, 17, 21, 22, 23, 24]. In fact, the increase in length is due, most often, to columnar metaplasia of the distal esophagus above the anatomic GEJ, although in some circumstances (ex. autoimmune gastritis, H.pylori infection), "cardia-type" mucosa, as described above, may increase in length by distal extension into the proximal corpus [25, 26]. In fact, a short segment of esophageal columnar mucosa, often only a few mm's in length, is a relatively common finding among patients who present for upper GI endoscopy in major hospital centers [1]. Although, theoretically, this represents BE, a new definition of BE has recently been adopted that refers only to those patients who have superimposed IM, characterized by the presence of goblet cells, since it is this latter type of epithelium that is a risk for neoplastic progression [10]. Unfortunately, this definition has serious limitations, given that IM is often focal and, thus, may be missed due to sampling error [27].

Etiology of Inflammation of GEJ Region ("GEJitis")
Gastroesophageal reflux disease and H.pylori infection are the major etiologic factors in the development of inflammation and IM of the GEJ region [1, 2]. In some individuals, both these etiologic agents may, in fact, act synergistically to cause inflammation [24, 28, 29]. The possibility that other, as yet unidentified, etiologic factors, such as NSAIDs, may be responsible for inflammation in the GEJ region needs to be considered, but alternative etiologies have not been investigated thoroughly.

GERD
GERD induced chronic carditis and/or columnar metaplasia of the distal esophagus occurs more often in white males, in early to mid adult life, and is more common in patients who consume alcohol and tobacco, compared to patients with H.pylori-induced carditis [2]. Endoscopically, the presence of a hiatus hernia, and/or the presence of an irregular, proximally located, Z-line relative to the anatomic GEJ, with or without signs of gastric type mucosa extending into the distal esophagus, and evidence of esophagitis, are endoscopic features in support of GERD as the etiology of inflammation in the GEJ region.

Pathologically, GERD patients often show only a mild degree of mononuclear inflammation within the lamina propria [16, 22]. However, acute inflammation, in the form of cryptitis, crypt abscesses or infiltration of the surface epithelium, either with or without surface erosion, may be present in some cases as well. Eosinophils are often a prominent component of the inflammatory infiltrate similar to that which occurs in the squamous epithelium in patients wit GERD-induced non-erosive esophagitis [22] (Figure 2). Eosinophils may infiltrate the columnar epithelium in the gastric cardia as well. A dense infiltrate of plasma cells, associated with reactive lymphoid follicles, are uncommon in GERD unless there is superimposed H.pylori infection. In a biopsy study by Wieczorek et al of 30 patients with known GERD induced inflammation, and 25 patients with known H.pylori-induced inflammation, of the GEJ region, GERD patients showed a significantly lower degree of plasma cells, neutrophils, and overall inflammation, but significantly more eosinophils, in comparison to H.pylori cases in which neutrophils, lymphocytes, plasma cells, and reactive lymphoid aggregates were not prevalent [30]. In addition to gastric-type mucinous epithelium and underlying mucous, or mixed mucous/oxyntic, glands, a variety of "hybrid" cells that contain both gastric and intestinal morphologic features, and "pseudogoblet" cells (representing distended foveolar cells), may be seen in biopsies from GEJ as well [31, 32, 33]. However, the specificity of those other cell types for esophageal columnar metaplasia versus native cardia epithelium is currently unknown.

In addition, the squamous mucosa in patients with GERD often shows reactive changes characterized by basal cell hyperplasia, elongation of the rete pegs and congestion of the lamina propria [2, 30]. Evidence of active esophagitis, characterized by eosinophilic and/or neutrophilic inflammation, may be present in more severe cases as well. In GERD patients who have a proximally displaced Z-line and, thus, columnar metaplasia of the distal esophagus, the lamina propria of the metaplastic columnar epithelium may look identical to that seen in the true gastric cardia. However, other features may be present that can help identify the epithelium as esophageal in origin. These include submucosal glands or ducts, ad multilayered epithelium (ME), and are discussed further below in the differential diagnosis section [34].

H.pylori
The clinical, endoscopic and pathologic features of H.pylori-induced chronic carditis are often distinct from those seen in GERD, but there is overlap [2, 30]. Clinically, H.pylori patients are usually older in age and have a more equal male to female ratio than GERD patients. Endoscopically, the Z-line usually approximates closely the proximal aspect of the gastric folds in non-GERD patients, and hiatus hernias are less common. In H.pylori patients, the cardia often shows similar pathologic features to that seen in H.pylori antritis, characterized by a variable increase in the amount of lymphocytes and plasma cells in the lamina propria, neutrophilic cryptitis, and reactive lymphoid aggregates [30] (Figure 3). In contrast to GERD, eosinophils are usually absent or only few in number. In addition, patients with H.pylori carditis, but without GERD, reveal normal non- inflamed esophageal squamous epithelium. Use of special stains is recommended for detection of H.pylori organisms [35, 36]. However, since H.pylori organisms do not normally colonize intestinal type epithelium, they may be few in number in biopsies that contain IM. An exception to this finding occurs in patients who have been previously treated with a proton pump inhibitor, in which case there is some evidence to suggest that the organisms migrate proximally in the stomach [37, 38].

Intestinal Metaplasia of the GEJ

Prevalence: Nearly one-third of patients who present for upper GI endoscopy and without endoscopic evidence of BE, reveal IM in the GEJ region [1, 39]. Not surprisingly, the changes of detecting IM has been shown to increase proportionally with the number of biopsies obtained at edoscopy [40]. In fact, there is some circumstantial evidence to suggest that the prevalence of IM is higher in patients who have longer lengths of mucosa composed of pure cardia type glands, and, therefore, may be higher in patients with GERD compared to those with H.pylori [16, 18, 21]. For instance, in a study by Chandrasoma et al of 959 patients in whom biopsies were obtained from the GEJ region, 70% of those who had between 1-2 cm of mucosa composed of mucous glands, or mixed mucous/oxyntic glands, in the GEJ region had IM in comparison to only 15% of patients who had <1 cm of "cardia-type" mucosa [18]. Similarly, several studies have shown an association between the finding of IM in biopsies of the GEJ region and male gender, white race, and higher patient age, features representative of a GERD clinical profile [41, 42, 43.]

Interestingly, IM detected in cardia biopsies from H.pylori patients is usually composed of a mixture of the incomplete and complete types in contrast to GERD-induced carditis and columnar metaplasia of the distal esophagus in which IM is usually of the incomplete type [3, 34, 42, 44]. The type and pattern of IM at the GEJ may explain the differences in dysplasia and cancer risk in patients with IM in the cardia related to H.pylori versus those with metaplastic columnar epithelium with IM in the distal esophagus related to GERD (see below)

Pathogenesis: It is widely believed that chronic inflammation is the underlying stimulus for the development of IM in the GEJ, regardless of the etiology [3, 22, 30, 45]. In general, most cardia biopsies contain, at minimum, a mild degree of chronic inflammation which is normally increased in comparison to the corpus or antrum, particularly in H.pylori negative patients [30]. In fact, a positive correlation was noted between the degree of chronic inflammation and the presence of IM in a biopsy study by Goldstein et al [46]. In GERD patients, it is commonly believed that chronic reflux leads to inflammation and ulceration of the native squamous epithelium, which if persistent, may lead to IM. Squamous epithelium converts first to cardia-type columnar epithelium composed of mucinous columnar epithelium and mucous or mixed mucous/oxyntic glands, prior to intestinalization [17, 32, 45]. Thus, with ongoing injury, and chronic inflammation, mucinous columnar epithelium converts to an intestinal phenotype as a result of secondary metaplastic reaction. Intestinal metaplasia related to GERD-induced metaplasia is usually of the incomplete type, which is composed of a mixture of acid-mucin containing goblet cells and gastric-type mucinous columnar epithelium. As mentioned below, the type and extent of IM is one of several factors that may help pathologists distinguish mucosal biopsies of the distal esophagus from the gastric cardia.

Unfortunately, the precise cell of origin of metaplastic columnar epithelium in the esophagus is unknown [2, 34]. However, possible sites of multipotential stem cells includes the basal layer of the native squamous epithelium, esophageal mucosa and/or submucosal glands and ducts, gastric cardia epithelium, congenital rests of gastric or intestinal epithelium in the esophagus and/or the subepithelium mesenchyme of the esophagus [34, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50]. Interestingly, recent studies suggest that metaplastic and dysplastic epithelium in the stomach may derived from bone marrow cells, but this theory has never been investigated in the esophagus [51]. Nevertheless, based on a series of experiments by Gillen et al and Li et al, it is now firmly believed that metaplastic epithelium in the esophagus is derived from cells that are, in fact, intrinsic to the esophagus rather than the stomach [50, 52]. There is also some evidence to suggest that esophageal mucosal ducts harbor stems that can differentiate into columnar epithelium [34, 47].

Some authors suggest that the squamous to columnar cell transition is the distal esophagus occurs via an intermediate, or transitional phase prior to intestinalization [17, 30, 34, 53]. In 1993, Shields et al reported the presence of a distinctive type of multilayered epithelium (ME) that shows morphologic and cytochemical characteristics of both squamous and columnar epithelium [54] (Figure 4). These investigators hypothesized that acid-induced ulceration of squamous epithelium exposes an as yet unidentified stem cell, which upon repeated acid damage, is stimulated to differentiate toward a columnar phenotype after passing through an intermediate ME phase. Multilayered epithelium is a biologically active epithelium that is phenotypically similar to fully developed BE [34]. This epithelium has been shown by Glickman et al to contain a high capacity for cell proliferation and differentiation [34]. It has also been shown to be highly associated with BE in a prospective biopsy study by Shields et al [53]. Multilayered epithelium was also recently shown to be strongly associated with GERD-induced inflammation of the GEJ region in a biopsy study by Wieczorek et al [30]. In their study, 30% of patients with GERD induced inflammation of the GEJ region had ME compared to only 4% (1 patient) of H.pylori patients (and this patient also had a hiatus hernia!). Thus, regardless of the putative role of ME in the development of columnar metaplasia in the esophagus, the presence of ME in a GEJ biopsy is considered highly specific for GERD and, thus, probably represents a specific histologic biomarker of GERD-induced metaplastic columnar epithelium in the distal esophagus [55]. Multilayered epithelium is usually detected at the SCJ and often in the vicinity of the openings of the submucosal gland ducts, observations supporting the theory that the submucosal gland duct epithelium may contain stems cells that give rise to metaplastic columnar epithelium [34, 53, 55].

In contrast, IM that develops in the true gastric cardia secondary to chronic H.pylori infection represents a columnar to columnar (goblet cell) metaplastic reaction. It is currently unclear whether GERD-induced inflammation in the true gastric cardia can lead to IM in this manner, although there is some evidence in favor of this mechanism [21, 16, 56]. Unfortunately, the pathogenesis of IM in the stomach has been poorly studied. Nevertheless, studies in the distal stomach have confirmed that chronic inflammation is a required precursor of IM and, thus, it is presumed that the same pathogenetic sequence occurred in the cardia of H.pylori infected patients, but this remains to be proven [2, 57]. In the stomach, the factors responsible for the conversion of mucinous columnar epithelium to intestinal-type epithelium are unknown. However, the cell or origin is presumed to reside in the deep foveolar or stem cell region of the gastric mucosa. In contrast to GERD-induced IM in the esophagus, IM in the cardia due to H.pylori infection is usually composed of a mixture of incomplete and complete-type IM.

Natural History: It is commonly believed that IM is the mucosal field upon which adenocarcinoma develops, both in the esophagus and stomach, regardless of the specific etiology [1, 41, 61]. In one study, 86% of adenocarcinomas that developed in the GEJ were associated with IM in adjacent, non-tumorous, mucosa [13]. With regard to adenocarcinomas of the GEJ region, evidence linking adenocarcinoma of the distal esophagus to GERD-induced IM is stronger than the data linking IM to the development of true cardia cancers [14, 59]. It is likely that most cancers develop through an IM-dysplasia-carcinoma sequence. Unfortunately, the true cancer risk for patients with IM in the distal esophagus secondary to GERD versus those with IM in true gastric cardia secondary to H.pylori is, essentially, unknown. In one study by Sharma et al, the dysplasia risk was shown to be greater in patients with short-segment BE versus those with IM of the gastric cardia only [14]. In that study, the prevalence rate of dysplasia in patients with short-segment BE versus those with carditis and IM was 11.3% versus 1.3%, with an incidence rates of 4.6% and 1.5%, respectively. In another study by the same lead author, the incidence of dysplasia in patients with short-segment BE was 5.7% per year [59]. However, in both of these studies, the risk of neoplasia was not defined specifically for patients with either ultrashort BE, or for those columnar metaplasia of the distal esophagus but without IM. Thus, although difficult to interpret, IM in metaplastic columnar epithelium in the distal esophagus (i.e. ultrashort or short-segment BE) probably has a higher likelihood of progression to malignancy in comparison to patients with I in the true gastric cardia related to H.pylori infection. Clearly, this issue requires further investigation.

Differential Diagnosis of H.pylori Versus GERD in GEJ Biopsies
As noted above, it is important to differentiate true carditis from columnar metaplasia of the distal esophagus, given that the etiology, pathogenesis, natural history, and possibly, risk of malignancy is different between these two pathologic conditions. Since, as mentioned above, it is difficult for endoscopists to know the precise anatomic location of a biopsy in the GEJ region (i.e. whether it is located above or below the anatomic GEJ) it is incumbent upon pathologists to help establish the precise location, and etiology of inflammation, of inflammatory conditions in mucosal biopsies from the GEJ region, whenever possible. Table #1 summarizes some of the features that can help differentiate these two disorders. Pathologically, the finding of esophageal mucosal or submucosal glands or ducts, or ME, in a biopsy from the GEJ is helpful to confirm that a particular biopsy is from the tubular esophagus, and if mucinous columnar epithelium is present, then it s reasonable to conclude that it is metaplastic in origin [2] (Figure 5). Other pathologic features may be helpful in this differential as well. For instance, in a recent study by Srivastava et al published in abstract form, a wide variety of objective morphologic parameters were evaluated in SCJ biopsies from the GEJ region in 20 patients with BE and 20 with carditis with IM [60]. In that study, the presence of crypt atrophy, crypt disarray, the presence and extent of incomplete IM (comprising >50% of crypts), the finding of squamous epithelium overlying crypts with IM, ME, esophageal gland/ducts, or the presence of hybrid glands (deep glands composed of IM in their most superficial aspect), were all features highly associated with BE in biopsies obtained from the GEJ region. In fact, squamous mucosa overlying crypts with IM, hybrid glands, and esophageal gland/ducts were each 100% specific for BE. Another recent study by that same research group confirmed that there is good agreement among GI pathologists for distinguishing BE from carditis with IM using the morphologic features outlined above (unpublished information).

The findings in esophageal squamous epithelium and in the reminder of the stomach (antrum, corpus) may also be helpful if biopsies are obtained from these locations. For instance, most, if not all cases, of H.pylori carditis also show H.pylori antritis and/or funditis [36, 36]. Conversely, the presence of active esophagitis in the squamous epithelium of the esophagus, combined with the finding of a normal antrum or corpus is strong evidence in favor of GERD as the cause of "GEJitis" [30]. As mentioned above, H.pylori stains should be performed on all biopsies from he GEJ region since positivity can help confirm the diagnosis readily.

Some studies have suggested that there are, in fact, differences in mucin histochemistry between esophageal columnar epithelium, or goblet cells, derived from squamous epithelium and either normal or inflamed gastric cardia type epithelium [32, 45, 61, 62, 63]. However, many of these studies suffer from limitations discussed previously in the introduction section of this review article. Nevertheless, in a recent well-controlled and rigidly defined mucosal biopsy study by Glickman et al, immunohistochemical expression of MUC-1 and MUC-6 were shown to be highly associated with BE-associated IM in contrast to IM in the cardia related to H.pylori infection [63]. In their study, combined MUC-1 and MUC-6 staining in goblet cells was 90% specific for intestinalized epithelium related to GERD, or BE. Unfortunately, the low sensitivity of these stains probably limits their use in clinical practice.

Basic mucin histochemical stains, such as PAS (neutral mucins), Alcian blue (acid mucins) and high-iron diamine [acidic (colonic-type) sulphomucins], have not been shown to be helpful in distinguishing esophageal-derived columnar epit4helium from the gastric cardia [32, 33, 64, 65]. Both non-goblet and goblet cells may be positive with any, or all, of these stains regardless of the site of origin. Some have advocated the use of Alcian blue to distinguish distended foveolar cells ("pseudogoblet" cells) (light staining) from true goblet cells (dark staining), regardless of heir location (Figure 6). Unfortunately, studies have documented that both of these cells types may stain either weakly or strongly with Alcian blue, regardless of whether they are from metaplastic columnar epithelium in the distal esophagus or from the proximal cardia [33, 64, 65]. In a study by Chen et al, high-iron diamine positivity in non-goblet columnar cells in mucosal biopsies from the GEJ region were shown to be highly associated with IM and, as such, the authors speculated that sulphomucin containing non-goblet cells are a specific marker of IM32. However, sulphomucin containing goblet cells were detected in both the esophagus and stomach in that study. The utility of other mucin histochemical markers such as DAS-1, CDX-2, HEP and CD10 have all been proposed as markers of IM in the esophagus, but non have been shown to be useful in this particular differential diagnosis [66, 67, 68, 69].

The pattern of cytokeratin 7 and 20 (CK7/20) immunostaining has recently been reported to help distinguish IM in the distal esophagus from IM in the true gastric cardia [8]. In 1999, Ormsby et al identified a distinctive type of CK7/20 immunostaining pattern in BE, consisting of diffuse strong CK7 staining of the surface and gland epithelium, and superficial weak columnar staining with CK20, and termed this the "BE CK7/20 staining pattern" [70]. The authors of that study found that the BE staining pattern was highly sensitive and specific for BE compared to gastric IM. Unfortunately, several other investigators have not been able to confirm the findings by Ormsy et al, and instead, have shown that the CK 7/20 immunostaining profile in biopsies from the GEJ or cardia often show a BE pattern as well [8, 45, 71, 72]. For instance, in a biopsy study by Glickman et al, patients with BE, both long and short segment types, as well as those with IM of the GEJ secondary to either H.pylori or GERD, showed a similar high degree of BE CK7/20 staining pattern (91% and 90%, respectively) [71]. In addition, there are many limitations to the use of CK7/20 immunostaining in evaluating biopsies from the GEJ region. These include differences in technical aspects of staining which create different staining results, observer variability in interpretation of a BE staining pattern, and variability depending on the types of samples utilized (biopsies versus resection specimens) [8]. For instance, in another inter-institutional observer variability study by Glickman et al, a different pattern of CK7/20 staining was observed in mucosal biopsies that were fixed in formalin versus those fixed in Hollandaise solutions [37]. In addition, investigators from two different institutions showed significant disagreement on recognition of a BE staining pattern. Another major limitation of the CK 7/20 immunostaining technique is that it requires an appropriate amount of mucosa with IM to be present in order for the staining reaction to be interpreted reliably, at this present in time, a feature that is rarely present in small mucosal biopsies from the GEJ region. Thus, use of CK7/20 immunostaining is not recommended on daily clinical practice to separate esophageal from gastric IM on small biopsies from the GEJ region.

References

  1. Spechler SJ. Intestinal metaplasia at the gastroesophageal junction. Gastroenterology 2004;124:567-75.

  2. Odze RD. Unraveling the mystery of the gastroesophageal junction: A pathologist's perspective. Am J Gastroenterol 2005;100:1853-67.

  3. Spechler SJ. The role of gastric carditis in metaplasia and neoplasia at the gastroesophageal junction. Gastroenterology 1999;117:218-28.

  4. Brown LM, Devesa SS. Epidemiologic trend in esophageal and gastric cancer in the United States. Surg Oncol Clin N Am 2002;11:235-56.

  5. Sharma P, McQuaid K, Dent J, et al. A critical reviwe of t4he diagnosis and management of Barrett's esophagus: The AGA Chicago Workshop. Gastroenterology 2004;127:310-30.

  6. Dinardi FG, Riddell RH. Esophagus. In: Stenberg S, ed. Histology for pathologists. New York: Raven Press, 1992:515-32.

  7. Shama P, Morales TG, Sampliner RE. Short segment Barrett's esophagus – the need for standardization of the definition and of endoscopic criteria. Am J Gastroenterol 1998;93:1033-6.

  8. Odze R. Cytokeratin 7/20 immunostaining: Barrett's oesophagus or gastric intestinal metaplasia? Lancet 2002;359:1711-3.

  9. Chandrasoma PT, Der R, Dalton P, et al. Distribution and significance of epithelial types in columnar lined esophagus. Am J Surg Pathol 2001;25(9):1188-93.

  10. Sampliner RE. Practice parameters committee ACG. Update guidelines for the diagnosis, surveillance and therapy of Barrett's esophagus. Am J Gastroenterol 2002;97:1888-95.

  11. Derdoy JJ, Bergwerk A, Cohn H, et al. The gastric cardia. To be or not to be? Am J Surg Pathol 2003;27(4):499-504.

  12. Goldblum JR. The significance and etiology of intestinal metaplasia of the esophagogastric junction. Ann Diag Pathol 2002;6(1):67-73.

  13. Cameron AJ, Souto EO, Smyrk TC. Small adenocarcinomas of the esophagogastric junction: association with intestinal metaplasia and dysplasia. Am J Gastroenterol 2002;97(6):1375-80.

  14. Sharma P, Weston AP, Morales T, et al. Relative risk of dysplasia for patients with intestinal metaplasia in the distal oesophagus and in the gastric cardia. Gut 2000;46:9-13.

  15. Kilgore SP, Ormsby AH, Gramlich TL, et al. Teh gastric cardia: Fact or fiction? Am J Gastroenterol 2000;95(4):921-4.

  16. Chandrasoma PT, Der R, Ma Y, et al. Histology of the gastroesophageal junction. An autopsy study. Am J Surg Pathol 2000;24(3):402-9.

  17. Glickman JN, Fox V, Antonioli DA, et al. Morphology of the cardia and significance of carditis in pediatric patients. Am J Surg Pathol 2002;26(8):1032-9.

  18. Chandrasoma PT, Der R, Ma Y, et al. Histologic classification of patients based on mapping biopsies of the gastroesophageal junction. Am J Surg Pathol 2003;27(7):929-36.

  19. Zhou H, Greco MA, Daum F, et al. Origin of cardiac mucosa: Ontogenic consideration. Ped Dev Pathol 2001;4:358-63.

  20. Park YS, Park HJ, Kang GH, et al. Histology of gastroesophageal junction in fetal and pediatric autopsy. Arch Pathol Lab Med 2003;127:451-5.

  21. Sarbia M, Donner A, Gabbert HE. Histopathology of the gastroesophageal junction. A study on 36 operations specimens. Am J Surg Pathol 2002;26(9):1207-12.

  22. Oberg S, Peters JH, DeMeester TR, et al. Inflammation and specialized intestinal metaplasia of cardiac mucosa is a manifestation of gastroesophageal reflux disease. Ann Surg 1997;226:522-32.

  23. Csendes A, Maluenda F, Braghetto I, et al. Location of the lower oesophageal sphincter and the squamous columnar mucosal junction in 109 healthy controls and 778 patients with different degrees of endoscopic oesophagitis. Gut 1999;34:21-7.

  24. Chandrasoma PT, Der R, Ma Y, et al. Histology of the gastroesophageal junction. Am J Surg Pathol 2000;24:402-9.

  25. Capella C, Fiocca R, Cornaggia M, et al. autoimmune gastritis. In Graham DY, Genta RM, Dixon MF, eds. Gastritis, 1st ed. Philadelphia : Lippincott Williams & Wilkins 1999;79-96.

  26. Karnes WE Jr, Samloff IM, Siurala M, et al. Positive serum antibody and negative tissue staining for Helicobacter pylori in subjects with atrophic body gastritis. Gastroenterology 1999;101:167-74.

  27. Guelrud M, Herrera I, Essenfel H, et al. Intestinal metaplasia of the gastric cardia: A prospective study with enhanced magnification endoscopy. Am J Gastroenterol 2002;97:584-9.

  28. Couverlard A, Cauvin JM, Goldfain D, et al. Cytokeratin immunoreactivity of intestinal metaplasia at normal oesophagogastric junction indicates its etiology. Gut 2001;49:761-6.

  29. Castro ML, Fachal C, Pineda JR, et al. Intestinal metaplasia at the esophagogastric junction. Prevalence and association in patients undergoing endoscopy. Gastroenterol Hepatol 2002;25(8):487-92.

  30. Wieczorek TJ, Wang HH, Antonioli DA, et al. Pathologic features of reflux and Helicobacter pylori-associated carditis. A comparative study. Am J Surg Pathol 2003;27(7):960-8.

  31. Bronner MP. Inflammatory disorders of the GI tract: Inflammatory disorders of the esophagus. In Odze, Goldblum, Crawford, eds. Surgical Pathology of the GI Tract, Liver, Biliary Tract and Pancreas. 1st ed. Philadelphia , PA : Saunders, 2004:121-142.

  32. Chen YY Wang HH, Antonioli DA, et al. Significance of acid-mucin-positive nongoblet columnar cells in the distal esophagus and gastroesophageal junction. Hum Pathol 1999;30(12):1488-95.

  33. Offner FA, Lewin KJ, Weinstein WM. Metaplastic columnar cells in Barrett's esophagus: a common and neglected cell type. Hum Pathol 1996;27:885-9.

  34. Glickman JN, Chen YY, Wang HH, et al. Phenotypic characteristics of a distinctive multilayered epithelium suggests that it is a precursor in the development of Barrett's esophagus. Am J Surg pathol 2001;25(5):569-78.

  35. Genta RM, Graham DY. Comparison of biopsy sites for t4he histopathologic diagnosis of Helicobacter pylori: A topographic study of H.pylori density and distribution. Gastrointest Endosc 1994;40:342-5.

  36. Genta RM, Huberman RM, Graham DY. The gastric cardia in Helicobacter pylori infection. Hum Pathol 1994;2:915-9.

  37. Kuipers EJ, Uyterlinde AM, Pena S, et al. Increased of Helicobacter pylori-associated corpus gastritis during acid suppressive therapy: implications for long-term safety. Am J Gastroenterol 1995;90:1401-6.

  38. Logan RPH, Walker MM, Misiewicz JJ, et al. Changes in the intragastric distribution of Helicobacter pylori during treatment wit omeprazole. Gut 1995;36:12-16.

  39. Spechler SJ, Zeroogian JM, Antonioli DA, et al. Prevalence of metaplasia at the gastro-oesophageal junction. Lance 1994;344:1533-6.

  40. Nandurkar S, Talley NJ , Martin CJ, et al. Short segment Barrett's oesophagus: prevalence, diagnosis and association. Gut 1997;40:710-5.

  41. Pera M. Trends in incidence and prevalence of specialized intestinal metaplasia, Barrett's esophagus, and adenocarcinoma of he gastroesophageal junction. World J Surg 2003;27:999-1008.

  42. Voutilainen M, Forkkilo M, Juhola M, et al. specialized columnar epithelium of the esophagogastric junction: Prevalence and associations. Am J Gastroenterol 1999;94;913-8.

  43. Ormsby AH, Kilgore SP, Goldblum JR, et al. teh location and frequency of intestinal metaplasia at the esophagogastric junction in 223 consecutive autopsies: Implications for patients treatment and preventive strategies in Barrett's esophagus. Mod Pathol 2000;13(6):14-20.

  44. Morales CP, Spechler SJ. Intestinal metaplasia at the gastroesophageal junction: Barrett's, bacteria, and biomarkers. Am Gastroenterol 2003;98(4):459-62.

  45. DeMeester SR, Wickramasinghe KS , Lord RVN, et al. Cytokeratin and DAS-1 immunostaining reveal similarities among cardia mucosa, CIM, and Barrett's esophagus. Am J Gastroenterol 2002;97(10):2514-23.

  46. Goldstein NS. Gastric cardia intestinal metaplasia: Biopsy follow-up of 85 patients. Mod Pathol 2000;13(10);1072-9.

  47. Adler RH. The lower esophagus lined by columnar epithelium: Its association with hiatal hernia, ulcer stricture, and tumor. J Thorac Cardiovasc Surg 1963;45:13-32.

  48. Bremner CH, Lynch CP, Ellis FH. Barrett's esophagus: Congenital or acquired: an experimental study of esophageal mucosal regeneration in the dog. Surgery 1970;68:209-16.

  49. Boch JA, Shields HM, Antonioli DA, et al. Distribution of cytokeratin markers in Barrett's specialized columnar epithelium. Gastroenterology 1997;112:760-5.

  50. Li H, Walsh TN, O'Dowd G, et al. Mechanisms of columnar metaplasia and squamous regeneration in experimental Barrett's esophagus. Surgery 1994;115:176-81.

  51. Houghton JM, Stoicov C, Nomura S, et al. Gastric cancer originating from bone marrow-derived cells. Science 2004;306:1568-71.

  52. Gillen P, Keeling P, Byrne PJ, et al. Experimental columnar metaplasia in the canine oesophagus. Br J Surg 1988;75:113-5.

  53. Shields H, Rosenberg SJ, Zwas FR, et al. Prospective evaluation of multilayered epithelium in Barrett's esophagus. Am J Gastroenterol 2001;96(12):3268-73.

  54. Shields HM, Zwas F, Antonioli DA, et al. Detection by scanning electron microscopy of a distinctive esophageal surface cell at the junction of squamous and Barrett's epithelium. Dig Dis Sci 1993;39:97-108.

  55. Glickman JN, Spechler SJ, Dineen T, et al. Multilayered epitheluim at the squamocolumnar junction is a histological marker for gastroesophageal reflux disease. Mod Pathol 2005;81(s1):103A.

  56. Chandrasome PT, Lokuhetty DM, DeMeester TR, et al. Definition of histopathologic changes in gastroesophageal reflux disease. Am J Surg Pathol 2000;24:344-51.

  57. Correa P, Felipe MI, Torrado J. Gastritis and gastric neoplasia. In: Graham DY, Genta RM, Dixon MF, eds. 1st ed. Philadelphia, PA: Lippincott Willaims and Wilkins, 1999:205-16.

  58. Ruol A, Parenti A, Zoninotto G, et al. Intestinal metaplasia is the probable common precursor of adenocarcinoma in Barrett's esophagus and adenocarcinoma of the gastric cardia.

  59. Sharma P, Morales TG, Battacharyya A, et al. Dysplasia in short-segment Barrett's esophagus: A prospective 3-year follow-up . Am J Gastroenterol 92(11):2012-6.

  60. Srivastava A, Glickman JN, Odze RD. Morphological parameters are useful in distinguishing Barrett's esophagus from carditis with intestinal metaplasia. Mod Pathol 2005;18(S1):119A.

  61. Rogge-Wolf C, Seldenrijk CA , Das KM, et al. Prevalence of mAB DAS01 positivity in biopsy specimens from the esophagogastric junction. Am J Gastroenterol 2002;97(12):297-85.

  62. Jovanovic I, Tzardi , Mouzas IA, et al. Changing pattern of cytokeratin 7 and 20 expression from normal epithelium to intestinal metaplasia of the gastric mucosa and gastroesophageal junction. Histol Histopathol 2002;17:445-54.

  63. Glickman NJ, Shahafaei A, Odze RD. Mucin core peptide expression can help differentiate Barrett's esophagus from intestinal metaplasia of the stomach. Am J Surg Pathol 2003;27(10):1357-65.

  64. Gottfried MR, McClave SA, Boyce HW: Incomplete intestinal metaplasia in the diagnosis of columnar lined esophagus (Barrett's esophagus). Am J Clin Pathol 1989;92:741-6.

  65. Peuchmaur M, Potet F, Goldfain D. Mucin histochemistry of the columnar epithelium of the esophagus (Barrett's esophagus): A prospective study. Clin Pathol 1984;37:607-10.

  66. Phillips RW, Frierson HF, Moskaluk CA. Cdx2as a marker of epithelial intestinal differentiation in the esophagus. Am J Surg Pathol 2003;27(11):1442-7.

  67. Chu PG, Jiang Z, Weiss LM. Hepatocyte antigen as a marker of intestinal metaplasia. Am J Surg Pathol 2003;27(7):952-9.

  68. Freund JN, Domon-Dell C, Kedinger M,e t al. Cdx1 and Cdx2 homeobox genes in the intestine. Biochem Cell Biol 1998;76:957-69.

  69. Chu PG, Ishizawa S, Wu E, et al. Hepatocyte antigen as a marker of hepatocellular carcinoma: An immunohistochemical comparison to CEA, CD10, and AFP. Am J Surg Pathol 2002;26:978-88.

  70. Ormsby AH, Goldblum JR, Rice W, et al. Cytokeratins subsets can reliably distinguish Barrett's esophagus from intestinal metaplasia of the stomach. Hum Pathol 1999;30:288-94.

  71. Glickman J, Wang HH, Das KM, et al. Phenotype of Barrett's esophagus and intestinal metaplasia of the distal esophagus and gastroesophageal junction. An immunohistochemical study of cytokeratins 7 and 20, Das-1 and 45 MI. Am J Surg Pathol 2001;25(1):87-94.

  72. Wallner B, Sylvan A, Janunger KG, et al. Immunohistochemical markers for Barrett's esophagus and associations to esophageal Z-line appearance. Scand J Gastroenterol 2001;36:910-5.

  73. Glickman JN, Ormsby AH, Gramlich TL, et al. Inter-institutional variability and effect of tissue fixative on the interpretation of Barrett's cytokeratin 7/20 immunoreactivity pattern in Barrett's esophagus. Hum Pathol 2005;36(1):58-65.