The term gastritis is often used somewhat loosely. Not infrequently, clinicians refer to a patient with epigastric pain or dyspepsia as having "gastritis," and radiologists sometimes diagnose "gastritis" on the basis of nonspecific radiographic changes, such as mucosal irregularity or swelling. However, gastritis is neither a clinical nor a radiologic diagnosis. Most often, it is a purely histologic diagnosis, made by the use of random or targeted endoscopic biopsies. Although some conditions that injure the gastric mucosa may result in inflammation, others do not. Thus, "gastritis," as the suffix -itis implies, is characterized by the presence of inflammatory cells. In contrast, gastropathy is used to refer to conditions in which inflammation is not a prominent feature, although there may be epithelial damage and regeneration. In gastropathies, there are often abnormalities of the mucosa visible at endoscopy, with or without histologic changes. Sometimes the mucosal appearance at endoscopy is typical or diagnostic of a particular condition. Therefore, in the case of gastritis, the diagnosis is always based on biopsy, whereas gastropathies are sometimes diagnosed on endoscopy alone, sometimes histology alone, and sometimes both. (Strictly speaking, the word gastropathy refers to any and all disorders of the stomach, but in this article, we use its more recent application). The categorization of entities into gastritides and gastropathies is an important concept and helps narrow the diagnostic possibilities in a given case. In this article, for the sake of simplicity, the term gastritis is often used generically to refer to both gastritides and gastropathies. One purpose of this article is to propose an approach to and a classification of pediatric gastritis that is pertinent to practicing pediatric endoscopists. In so doing, we hope to focus attention on this little-studied area and to facilitate diagnosis of the various entities that comprise the classification. Further, by offering practical advice for endoscopists regarding endoscopic landmarks, terminology, and acquisition of biopsy specimens, we hope to facilitate consistency of communication among endoscopists and between endoscopists and pathologists. This, it is hoped, will lead to further work in pediatric gastritis and to improved diagnostic accuracy. HIGH-YIELD GASTROSCOPY Once a child is scheduled for diagnostic upper gastrointestinal endoscopy, the endoscopist has an obligation to maximize the diagnostic potential of the procedure. Central to this must be the recognition that the ability to make a diagnosis based on macroscopic appearances alone is limited (1-5). For example, gastric mucosa that appears normal at endoscopy may harbor marked inflammation visible on microscopic analysis. The converse is also true-that is, a markedly red mucosa at endoscopy may be normal or may result from contact with the endoscope. Redness may reflect underlying microvascular congestion without the presence of inflammatory cells. Thus, there is often poor correlation between endoscopic and histologic findings. It follows, therefore, that multiple gastric biopsies must be an integral part of a proper examination of the gastric mucosa (1-5). In children, gastritis remains underrecognized and poorly characterized because of the flawed tendency to rely on macroscopic appearances at endoscopy, and an apparent reluctance to perform multiple gastric biopsies. In evaluating specimens obtained in such biopsies, collaboration with a pathologist with special expertise in gastrointestinal histopathology is of inestimable value. A high yield of accurate diagnoses, including the definitive ruling out of mucosal disease, largely depends on the endoscopist's care and attention to detail in the following areas. Endoscopic Landmarks and Histologic Zones In reporting endoscopic appearances, the endoscopist is describing the gross pathologic appearance of disease. Gastric disorders often have a predilection for a particular area or zone of the stomach, and it is therefore important to identify endoscopic landmarks accurately. Inaccurate landmark identification may result in poorly targeted biopsies and erroneous or missed diagnoses. The largest region of the stomach is the gastric body or corpus, characterized at endoscopy by thick mucosal folds or rugae. The body or corpus extends distally to the incisura or angulus on the lesser curvature. The fundus is the dome-shaped area immediately above the gastric body and abutting the diaphragm. Both corpus and fundus are comprised of oxyntic mucosa. The gastric antrum occupies the lower quarter or third of the stomach. When the stomach is distended with air to allow visualization, the antrum is seen to begin on the greater curvature where folds of the gastric body end. The antrum ends at the pylorus and consists of mucus glands. The gastric cardia comprises a short zone-in children, extending in length from a few millimeters to perhaps one centimeter from immediately below the Z-line-composed of mucus glands. Figure 1 shows anatomic landmarks of the stomach.FIG. 1: Anatomic regions of the stomach. Reprinted with permission from Rudolph AM, Hoffman JIE, Rudolph CD, eds.Rudolph's Pediatrics. 20th ed. 1996, Appleton & Lange, Stanford, CT .The different histologic zones of the stomach correspond to the different gross anatomic zones thereof, although not precisely, because there is always some over-lap and interdigitation of histologic zones at areas of transition. Therefore, the term transitional mucosa applies to these. Gastric landmarks and their histologic correlates are described in more detail elsewhere (6-8). Although the number of inflammatory cells in the gastric mucosa of normal children has not been extensively studied, there are some published data, summarized elsewhere (8). Terminology Use of nonstandardized terminology or interpretive terms results in poor interobserver reliability of findings and uninterpretable endoscopy reports. It is therefore more helpful to record endoscopic findings in factual, descriptive terms. The endoscopist should report only what he or she sees in terminology that is standard and unambiguous. Jargon, "-itis" terms, and interpretive terms should be avoided in the factual or objective part of the endoscopy report. For example, use of the term "gastritis" could mean anything from erythema to distinct erosions. If the mucosa is red, it should be documented as red or erythematous (mild, moderate, intense, or hemorrhagic) and not as antritis or gastritis, because inflammation may not be present (2). After all, the human eye is not a microscope. An erosion is a mucosal break that does not penetrate the muscularis mucosae, whereas an ulcer extends through the muscularis into the submucosa. Endoscopists cannot accurately determine depth of lesions, but there are some clues. Erosions are often multiple and usually have white bases. Each erosion usually is surrounded by a ring of erythema. When erosions have recently bled, their bases may be black. Hemorrhage refers to a bright, shiny red appearance of the mucosa in patches, streaks, or discrete petechiae, not associated with a visible mucosal break. Although the term submucosal hemorrhage sometimes is used, endoscopists cannot see through the muscularis mucosae, and subepithelial hemorrhage is therefore preferable, to allow for varying depths of hemorrhage. Other confusing terms used for subepithelial hemorrhage (and best avoided) are acute gastritis, hemorrhagic gastritis (inflammation is usually absent from hemorrhagic lesions), or hemorrhagic erosion (usually, no is erosion present) (9). If gastric rugae are large, accurate terms of description are thick folds or swollen folds, not edematous folds or hypertrophic folds, because edema and hypertrophy are histologic, not endoscopic, findings. It is also important to describe nodules in the stomach accurately. For example, a nodule or patch of nodules may be seen occasionally, especially at the antrum-body junction; these may represent a prominent areae gastricae(10) and usually are unimpressive. In contrast, a continuous diffuse carpet of nodules spread throughout the antrum is characteristic of Helicobacter pylori infection, past or present (11,12) (Fig. 2A). These nodules are described further in the section on H. pylori. The nodules of chronic varioliform gastritis are again different. They are larger, discrete, more raised, often have an umbilicated central erosion or shallow ulcer, and predominate in the gastric fundus and body (13-15) (Fig. 3A). Large nodules, more like blebs, may occur in the proximal stomach with cytomegalovirus (CMV) gastritis (2). Similar nodules may occur in the antrum in eosinophilic gastritis. Thus, it is best for the endoscopist to describe accurately what is seen and to offer an interpretation or differential diagnosis only at the end of the endoscopy report, based on the factual findings.FIG. 2: (A) Endoscopic view to show the gastric antrum in Helicobacter pylori infection, with the pylorus in the distance. Once biopsy samples have been taken, oozing blood acts as a vital stain "hematochromoendoscopy," making visible a confluent carpet of nodules. This is a typical appearance, but it is not diagnostic of current H. pylori infection, because the nodules may remain for months or years after successful eradication of H. pylori. Nodules are not present in the corpus or fundus in H. pylori infection. (B) Helicobacter pylori gastritis. This gastric antral specimen shows two hyperplastic lymphoid follicles. Between these, there is chronic and active inflammation, particularly in the superficial mucosa. The inflammation comprises lymphocytes, plasma cells, and neutrophils. Hematoxylin and eosin; magnification, ×40.FIG. 3: (A) Endoscopic view of the proximal corpus to show the striking, large, "juicy" nodules typical of chronic varioliform gastritis. Typically, these are present in the corpus and fundus and not in the antrum. (B) Chronic varioliform gastritis. The specimen is of the gastric body, which is mildly edematous and inflamed with a mixture of leukocytes, plasma cells, and eosinophils. Note in the superficial subepithelial zone, there is a layer of collagen deposition (arrows) indicative of collagenous gastritis. This is not a constant finding in chronic varioliform gastritis. Hematoxylin and eosin; magnification, ×100.Biopsy Sampling Different disorders may have a predilection for one topographic zone or another of the stomach. Sometimes, the same agent causes different patterns of injury in different populations, (e.g., H. pylori) (16-19). Sometimes there is disease in more than one zone of the stomach, as in H. pylori infection, Crohn's disease, eosinophilic gastritis, atrophic gastritis, and CMV, as will be described. Thus, biopsy samples must be obtained from different topographic zones of the stomach (1,8,9,19). Even with careful handling of biopsies by trained personnel, endoscopic biopsy specimens may sustain crush or other artifacts. Thus, when biopsy is indicated, at least two samples should be obtained from a particular lesion or zone of the stomach. Although the optimum number of biopsy samples has yet to be determined, when mucosa is normal-appearing, our practice is to take at least two samples from the prepyloric or midantrum and two from the greater curve of the midbody. Others take two from the antrum-body transition zone of the lesser curve (a zone in which inflammation and metaplasia occur in adults) (1). In addition, we and others obtain specimens of the gastric cardia, immediately below the Z-line, when H. pylori infection, Barrett's metaplasia, carditis (caused by gastroesophageal reflux or H. pylori) (19-23), or mucosa-associated lymphoid tissue (MALT) lymphoma (24) may be present. When endoscopic findings are puzzling, or a lesion is present, more specimens should be taken randomly and from the lesion or edge thereof. The size of the biopsy specimens is also important. Samples obtained with "pediatric" forceps are often of little value, because they are tiny and difficult to mount. Thus, the quality and quantity of tissue are often insufficient. In most children more than 2 or 3 months of age, an endoscope with a 2.8-mm biopsy channel usually can be used (e.g., Model XQ20; Olympus, Tokyo, Japan) and samples obtained with these forceps often are adequate, if several are taken and care is taken with mounting and handling. In contrast, each specimen obtained with "jumbo" or large-cup forceps provides at least twice or three times the amount of mucosa for diagnosis, and these can often be used in older children. CLASSIFICATION We have previously described a grading system for pediatric gastritis (11), but to the best of our knowledge, other than an earlier pathologic description of pediatric gastritis from our group (8), there is no published classification of pediatric gastritis. Our purpose is to fill this void by proposing a classification geared to the needs of pediatric endoscopists. Further, by proposing a conceptual framework of gastritis and gastropathy and describing the specific entities encountered in the pediatric stomach and their differential diagnoses, we hope to make the process of accurate diagnosis easier for the clinician. No classification of gastritis satisfies everyone, because the published classifications have different objectives. Some are a glossary of appearances (25); others list clinical disorders (2). The Sydney system is essentially a checklist to aid the pathologist in reviewing biopsy specimens and in uniformly reporting findings, by use of a visual analog scale (18). Thus it is aptly named a system rather than a classification (18,25-27). Its usefulness to pediatric endoscopists is limited, because its major focus is grading the severity of chronic gastritis, atrophy, and intestinal metaplasia, which are of concern more in adults than children. Furthermore, it does not integrate histopathology with endoscopic appearance, and it does not classify noninflammatory conditions (18). We have chosen to classify mucosal disorders of the stomach in children primarily by their endoscopic appearances. In this system, gastritis is classified into two groups: erosive and/or hemorrhagic gastritis or gastropathy, and nonerosive gastritis or gastropathy (Table 1). Although some disorders can be either erosive or nonerosive, each is classified by its most common clinical manifestation. The disorders in each group are placed in approximate sequence of their prevalence in our practice. Each disorder is then described by cause and any distinctive clinical, endoscopic, and histologic features. We believe that the advantages of this approach are simplicity and ease of use for the practicing pediatric endoscopist. Based on this approach, a diagnosis or differential diagnosis may be made with some certainty at the time of endoscopy; however, for most disorders, confirmation of the initial impression, or definitive diagnosis is still dependent on biopsies, and therefore on an active dialogue and close collaboration with a pathologist.TABLE 1: Classification of gastritis and gastrophy in childrenaErosive and/or Hemorrhagic Gastritis or Gastropathy Most of these entities are diagnosed endoscopically, usually in patients presenting with gastrointestinal bleeding. Because inflammation is not a feature of most hemorrhagic lesions, most conditions in this category are gastropathies. Biopsies are usually not required from erosive or hemorrhagic lesions, where the diagnosis is clear from the clinical context and endoscopic findings. However, there are gastritides not in the erosive or hemorrhagic category that may have erosions or hemorrhagic lesions, and in these cases, biopsies are essential for diagnosis (e.g., H. pylori, Crohn's disease, CMV, and allergic gastritis). Therefore, whenever there is the slightest doubt, biopsy samples should be taken. "Stress" Gastropathy This usually occurs within 24 hours of the onset of critical illness in which physiological stress is present, such as shock, hypoxemia, acidosis, sepsis, burns, major surgery, multiple organ system failure, or head injury. These stressors cause reduction of gastric blood flow with subsequent mucosal ischemia (28) and breakdown of mucosal defenses (29). Gastric acid is important in the pathogenesis of stress erosions, but actual hypersecretion is only seen in sepsis, and in trauma affecting the central nervous system. Risk factors for hemorrhage include gastric hypersecretion, mechanical ventilation, and use of high-dose corticosteroids (29,30). Stress erosions typically are asymptomatic, multiple, and do not perforate, but when they produce symptoms, they do so with overt upper gastrointestinal hemorrhage. Newborns and infants appear to be more prone to perforations (31). Early stress lesions predominate in the fundus and proximal body, later spreading to the antrum to produce a diffuse erosive and hemorrhagic appearance. Antral involvement alone is uncommon. Neonatal Gastropathies In clinical practice, gastric mucosal disorders are seldom identified in the newborn, even in sick preterm infants. Perhaps this is in part because endoscopy is seldom indicated or because of a greater reluctance to perform endoscopy in infants. Nevertheless, a high prevalence of hemorrhagic gastropathy has been reported in sick neonates in an intensive care unit in whom no upper gastrointestinal symptoms or signs were present and who underwent endoscopy under a research protocol (32). Of note is that newborns with (33) and without (32) upper gastrointestinal symptoms seem to have a high prevalence of "esophagitis" associated with gastropathy. Hemorrhagic gastropathy has also been reported in otherwise healthy full-term infants (34) presenting with severe upper gastrointestinal hemorrhage-in one case, an antenatal hemorrhage (35). An unusual gastropathy may occur in infants with congenital heart disease receiving prolonged infusions of prostaglandin E to maintain patency of the ductus arteriosus. This consists of antral mucosal thickening or a focal mass caused by foveolar cell hyperplasia, manifesting as gastric outlet obstruction (36). This entity has also been described in a 6-week-old infant, who received no medications (37). Traumatic Gastropathy Forceful retching or vomiting produces typical subepithelial hemorrhages in the fundus and proximal body of the stomach. It is caused by "knuckling" or trapping of the proximal stomach in the distal esophagus, resulting in vascular congestion, and is also known as prolapse gastropathy(38,39). Mallory-Weiss tears immediately above or below the gastroesophageal junction also may occur. Although prolapse gastropathy and tears tend to resolve quickly, they can result in significant blood loss. By a similar mechanism of trauma, linear erosions may occur in the herniated gastric mucosa of patients with large hiatal hernias, resulting in anemia from chronic blood loss (40). Suction through nasogastric tubes, ingestion of foreign bodies, and endoscopic procedures such as diathermy (41,42) are common causes of subepithelial hemorrhages and focal erosions. Aspirin and Other Nonsteroidal Antiinflammatory Drugs Nonsteroidal antiinflammatory drugs (NSAIDs) produce mucosal injury by local irritant and systemic effects (43,44). Although delayed-release and enteric-coated NSAIDs produce less acute, superficial mucosal damage, they still cause ulcers and the complications thereof. Even a single dose of aspirin may cause petechial hemorrhages in the stomach within a few hours and erosions within 24 hours (45). However, early lesions usually are of little clinical significance and are not predictive of clinically significant ulcer formation (43). Nevertheless, in severe erosive gastropathy caused by NSAIDs, ulcers may bleed and/or perforate. Lesions caused by NSAIDs are more commonly gastric than duodenal, occurring more typically in the gastric antrum than in the body. In children, hemorrhagic antral gastropathy and ulceration of the incisura are the typical NSAID lesions. Occasionally, more extensive gastric involvement occurs, as does duodenal ulceration. Bleeding from such lesions after ingestion of NSAIDs in children has been well documented (46-49). In one study, 75% of children with juvenile rheumatoid arthritis who had taken one or more NSAIDs for more than 2 months had endoscopic evidence of gastropathy, antral erosions, or ulcers (48). Of these, 64% had anemia and abdominal pain. Although NSAID-induced gastropathy does not require the presence of H. pylori for its development (50), in adults taking NSAIDs, there are conflicting data regarding whether H. pylori eradication improves the rate of healing of peptic ulcers. Nevertheless, there seems to be agreement that H. pylori-infected people who have a history of peptic ulcer and who require NSAID therapy should have treatment directed at eradication of the organism (51). Other Drugs Although many drugs may cause nonulcer dyspepsia, erosive or hemorrhagic gastropathies have been described with valproic acid, dexamethasone, chemotherapeutic agents, alcohol, and potassium chloride (52-60). Portal Hypertensive Gastropathy This congestive gastropathy occurs frequently in children with intrahepatic or extrahepatic causes of portal hypertension (61). The endoscopic findings in portal hypertensive gastropathy (PHG) vary from mild gastropathy with a mosaic pattern of 2- to 5-mm erythematous patches separated by a fine white lattice, to severe gastropathy typified by the presence of cherry red spots or even a confluent hemorrhagic appearance (62,63). The mosaic pattern is specific for PHG and was not found at endoscopy in any of 500 children without liver disease (61). In adults, congestive gastropathy is more frequently associated with large gastroesophageal varices than with esophageal varices alone (64), and previous sclerotherapy of esophageal varices may exacerbate PHG (64,65). The diagnosis of PHG is made at endoscopy. Biopsy is seldom indicated and is potentially dangerous. The histologic findings in PHG are ectasia of mucosal capillaries and venules, and submucosal venous dilatation (63). Uremic Gastropathy In acute renal failure, gastropathy is caused by acute physiological stress, rather than renal failure itself. In adults with chronic renal failure receiving hemodialysis, gastroduodenal lesions occur in up to 67%, with the predominant lesion, antral gastropathy, in up to 50%. The presence and severity of gastroduodenal disease are not consistently related to the degree of hypergastrinemia, nor to that of chronic renal failure (66). Vascular ectasia in the stomach is an important cause of bleeding in chronic renal failure (67). Chronic Varioliform Gastritis Also known as chronic erosive gastritis, chronic varioliform gastritis is an uncommon disorder of unknown cause reported largely in middle-aged and elderly men (13). It has been reported in a few children (14,15,68,69), with variable combinations of upper gastrointestinal symptoms, anemia, protein-losing enteropathy, peripheral eosinophilia, and elevated serum immunoglobulin (Ig) E levels (68,69). Symptoms arise insidiously and often become subacute or chronic. Most striking endoscopically are the innumerable prominent nodules in the fundus and proximal body of the stomach (Fig. 3A). In children, the antrum is less often involved. Typically, the gastric rugae are irregularly thickened, with nodules located on the crests of the folds. The nodules sometimes have an umbilicated central crater or erosion; the lesions are said to resemble the skin lesions of chicken pox-therefore, the name. Histologic features include edema, foveolar hyperplasia, active chronic inflammation, and eosinophilic infiltrates. When present, focal superficial subepithelial collagen deposition may represent fibrosis at points of previous surface erosions (Fig. 3B). In some cases, there is an infiltrate of intraepithelial lymphocytes in the surface and foveolar epithelium, suggesting that chronic varioliform gastritis may be one cause of lymphocytic gastritis (70). We have observed variable degrees of collagen deposition with active inflammation and gland atrophy in three adolescents (see Collagenous Gastritis). Bile Gastropathy This is also known as alkaline gastropathy, or gastropathy caused by duodenogastric reflux. It has been well documented in the postoperative stomach (71) and in the intact stomach (72) in reports that are mainly in the adult literature. Typical endoscopic features include "beefy" redness or erythema and, occasionally, erosions. Despite this, there is very little cellular infiltrate in the lamina propria. The main histologic features are foveolar hyperplasia, occasionally with a corkscrew appearance; lamina propria edema; and venous congestion. These changes constitute the entity of so-called reactive gastropathy(9). After surgery, they are found more commonly in the stomach than at the stoma. Other features include stomal erosions, lipid islands, and mucosal cysts; the latter are sometimes grossly visible and are known as gastritis cystica profunda or polyposa. Some studies report a high prevalence of intestinal metaplasia, although this may reflect sampling from the stromal region, which normally reflects a mosaic of gastric and intestinal mucosa. Although the postoperative syndrome of postprandial pain, nausea, and vomiting often is ascribed to duodenogastric reflux, there is poor correlation between symptoms and endoscopic and histologic findings. Intermittent reflux of bile may occur even in normal people (73), therefore, the mere presence of bile in the stomach at endoscopy does not carry clinical significance. Fortunately, currently there are hardly any indications for partial gastrectomy in children, and pyloroplasty in children (74) is seldom recognized to be attended by the described problems. Henoch-Schönlein Gastritis Endoscopy is seldom required for diagnosis of this condition, but when the diagnosis is uncertain, endoscopy may be helpful, showing a hemorrhagic and erosive picture typical of this disorder. In our experience and that of others (75,76) in children undergoing upper gastrointestinal endoscopy for symptoms including epigastric pain, hematemesis, and vomiting, the antral mucosa is erythematous and swollen with raised blebs 3 to 5 mm in diameter. On the blebs are punctate hemorrhages and, often, central erosions or ulcer with a yellow base. Similar lesions are often present in the duodenum. Although gastric mucosal specimens usually are too superficial to show typical histologic changes, they may show a leukocytoclastic vasculitis similar to that seen in the skin (75). Corrosive Gastropathy The most commonly ingested substances that injure the stomach are acids, iron, and strong alkalis. The latter predominantly involve the esophagus but occasionally involve the stomach. When gastric injury occurs, the prepyloric area is particularly vulnerable (77,78), probably because of pylorospasm and pooling of secretions. The presence of food may limit the degree of injury. Endoscopic findings run the gamut from mild friability and erythema to necrosis, ulcers, exudates, hemorrhage, and, rarely, perforation. Chronic cicatrization is relatively rare, and may take several months to become apparent. Iron poisoning, especially with ferrous sulphate, is common in children in some areas of the world and may cause corrosive gastropathy with stricture (79). Therapeutic administration of oral ferrous sulfate may cause mild endoscopic abnormalities in the stomach that are of uncertain clinical significance (80). Ingestion of pine oil cleaner may also cause cause gastric injury (81). Exercise-Induced Gastropathy or Gastritis This condition is well recognized in runners, usually presenting with blood loss anemia, with or without upper gastrointestinal symptoms. Erosive gastropathy has been described (82), as has nonerosive gastritis (83), the latter with acute inflammation on biopsy. Postulated mechanisms include splanchnic ischemia and repeated jarring of the organ. Radiation Gastropathy This condition is rare but has been associated with massive abdominal irradiation in patients with malignancy, causing erosions or ulcers, particularly in the gastric antrum and prepyloric regions (84). Fibrosis and stricture formation may lead to gastric outlet obstruction. Nonerosive Gastritis or Gastropathy In nonerosive gastritis, there is usually a poor correlation between endoscopic appearance and histologic findings-that is to say, the diagnosis is almost always purely histologic. An exception is the nodular antrum of H. pylori-associated ulcer disease in children; however, nodularity persists even after eradication of H. pylori. Therefore, in this case the diagnosis is endoscopic and histologic. Furthermore, nodularity is not always present, and the diagnosis, again, ultimately depends on histology. Some of the entities in this section may also manifest endoscopically as an erosive gastropathy or gastritis but are included here because they more commonly occur without erosions. Lymphocytic gastritis is a type of gastritis deserving of special mention. It may be seen in disorders as apparently diverse as celiac disease, CMV gastritis, Menetrier's disease, H. pylori infection, and chronic varioliform gastritis. Because ours is an endoscopic classification, lymphocytic gastritis is mentioned under each of those disease entities. Nonspecific Gastritis In our experience, a significant number of children have chronic gastritis for which no cause can be identified (8). In these cases, the inflammation is chronic, with lymphoid and plasma cells predominant, more focal than diffuse within the biopsy, and usually superficial. Although it appears to be more prevalent in the antrum than the corpus, thi