- Human pathology

Home > E. Pathology by systems > Digestive system > Stomach > Sydney System for the classification of chronic gastritis

Sydney System for the classification of chronic gastritis

Thursday 23 February 2012

Sydney system grading of chronic gastritis

The Sydney System for the classification of gastritis emphasized the importance of combining topographical, morphological, and etiological information into a schema that would help to generate reproducible and clinically useful diagnoses.

The spectrum of gastritis encompasses several groups of nosological entities that for convenience can be divided into three broad categories: acute, chronic, and special (or “distinctive”) forms.

Sydney system of grading

Feature Definition Grading Guidelines
Chronic inflammation Increased lymphocytes and plasma cells in the lamina propria Mild, moderate, or severe increase in density
Activity Neutrophilic infiltrates of the lamina propria, pits, or surface epithelium Less than one third of pits and surface infiltrated = mild; one third to two thirds = moderate; more than two thirds = severe
Atrophy Loss of specialized glands from either antrum or corpus Mild, moderate, or severe loss
Intestinal metaplasia Intestinal metaplasia of the epithelium Less than one third of mucosa involved = mild; one third to two thirds = moderate; more than two thirds = severe
Helicobacter pylori H. pylori density Scattered organisms covering less than one third of the surface = mild colonization; large clusters or a continuous layer over two thirds of surface = severe; intermediate numbers = moderate colonization

Biopsy Sites

Mapping studies in which multiple biopsy specimens have been taken from H. pylori-positive subjects confirm that the careful examination of four specimens (two antral and two corpus) has a high probability of establishing the correct H. pylori status.

Corpus biopsies are particularly valuable for yielding positive results after treatment, especially where proton pump inhibitors have been used.

Under these circumstances, organisms may become rare or disappear from the antrum but remain in the oxyntic mucosa, which may also develop cystic dilatations with hypertrophy of the parietal cells.

Furthermore, biopsy specimens from the corpus are essential to establish the pattern of gastritis, which has important implications for the risk of associated diseases.

Whereas biopsy specimens from the antrum and corpus are adequate, therefore, to establish both the H. pylori status and the “background” level and distribution of gastritis, it is self-evident that additional biopsy specimens are required from any lesions that may be present and for mapping the distribution and extent of intestinal metaplasia and dysplasia.

However, maximal degrees of gastric mucosal atrophy and intestinal metaplasia are consistently found in the region of the incisura angularis, which is also the site most likely to reveal premalignant dysplasia.

Thus, specimens from the antrum and corpus should be supplemented with a fifth or more biopsy specimens from the incisura.

The choice of additional sites from which biopsy specimens may be taken is dependent on the local epidemiologic conditions with respect to the types of gastritis and the incidence of gastric carcinoma.

Biopsy specimens from the three regions should be separately identifiable when submitted to the laboratory.

Proper orientation is indispensable for optimal histological evaluation, but it is difficult to achieve with small pinch biopsies.

It may be accomplished with larger biopsies either in the endoscopy suite when biopsy specimens are collected or in the histopathology laboratory at the time of embedding.

Communication Between the Endoscopist and Pathologist

Full and accurate clinicopathologic correlation in gastritis is consistently achievable only when the pathologist is aware of biopsy locations and of relevant endoscopic and clinical observations.

Endoscopic information should include a brief narrative or a diagram describing focal lesions (e.g., thickened folds, polyps, masses, erosions, or ulcers) and abnormal-looking areas.

Specific questions (e.g., Is there atrophy, intestinal metaplasia, ectopic mucosa?) may also help to alert the pathologist about the possibility of a process that might not be searched for otherwise.

Endoscopic pictures forwarded warded to the pathologist can be particularly helpful.

Clinical information furnished to the pathologist should include the medications taken by the patients within the last 2 months, particularly nonsteroidal anti-inflammatory drugs, proton pump inhibitors, antibiotics, and nonprescription antiacids.

Other pertinent aspects of the medical history should also be included.

Special Stains

In addition to hematoxylin and eosin (H & E), many laboratories routinely undertake a special stain for H. pylori.

This practice encourages proper assessment and may be more cost effective than subsequent requests for an extra stain.

The choice of stain, for example, modified Giemsa, Warthin-Starry, or the new Genta stain, is a matter of local preference, but the use of a special stain is strongly recommended, particularly when H & E fails to reveal organisms in a biopsy specimen with chronic active inflammation.

Immunostains are also available for the demonstration of H. pylori and may be particularly useful in detecting coccoid forms.

Thus, although many positive cases can be recognized in a good H & E stain, careful examination of a special stain is essential before declaring an inflamed biopsy specimen histologically negative for H. pylori.

Many laboratories routinely employ an Alcian-blue (AB, pH 2.5)/periodic acid Schiff (PAS) stain on gastric specimens to demonstrate intestinal metaplasia.


A. For optimal assessment, five biopsy specimens are taken, two from the antrum within 2 to 3 cm from the pylorus, one from the distal lesser curvature, and the other from the distal greater curvature, two from the corpus about 8 cm from the cardia (one from the lesser and the other from the greater curvature), and one from the incisura angularis.

B. Samples from antrum, corpus, and incisura angularis should be separately identifiable.

C. Transmission of information to the pathologist about the patient’s endoscopic findings, clinical history, and biopsy sites is essential for successful clinicopathologic correlation in gastritis.

D. A special stain for H. pylori should be carried out before declaring an inflamed biopsy specimen negative.

E. An AB/PAS stain will facilitate the recognition of intestinal metaplasia.


The real value of classifying chronic gastritis lies in the relationship between certain patterns of inflammation and atrophy and their disease associations.

In some instances, these patterns are sufficiently consistent for them to be predictive of peptic ulcer or gastric cancer risk.

To translate the histopathological observations into well-defined topographic patterns or for comparison purposes, it is highly desirable to grade each relevant feature using a standardized reproducible scale.

Grading the morphological variables is useful in the evaluation of H. pylori gastritis and its related forms (nonatrophic and atrophic gastritides).

Although an assessment of the intensity of the inflammatory components may be useful in some of the special forms (e.g., non-H. pylori infectious and lymphocytic gastritis), a formal report of the graded variables is not considered necessary when reporting chemical, radiation, granulomatous, or eosinophilic gastritis.

Graded Variables

H. pylori density

For clinical management purposes, the most important information is whether Helicobacter is present.

Variations in H. pylori density may have a bearing on disease associations and have epidemiological importance.

Difficulties applying the grading guidelines are particularly apparent in biopsy specimens in which portions of the gastric mucosa show intestinal metaplasia, which is usually not colonized by H. pylori.

Observers have been uncertain about whether to grade the bacterial density on the gastric epithelium alone (ignoring the presence of the adjacent metaplasia) or to average the density of the bacteria over the length of the entire specimen.

SOme authiors recommend the first approach, which limits the evaluation of H. pylori density to the areas where it normally resides.

Polymorphonuclear neutrophil activity

Notwithstanding the problems surrounding the term activity, it was considered worthwhile to retain it, as in chronic active gastritis, and to grade its severity.

In this context, it provides a useful short-hand for “the presence of neutrophil polymorphs in a background of chronic inflammation.”

As such, it is a measure of continuing acute inflammation; given the likely role of neutrophil-derived reactive oxygen species and proteases, neutrophil “activity” is likely to be linked to tissue damage.

Chronic inflammation in the absence of neutrophils is also “active” in the sense that cytotoxic T-lymphocytes and other cell effectors may play a role in tissue damage, and operate in glandular destruction in some patterns of gastritis.

Neutrophil activity is an almost universal phenomenon in H. pylori gastritis.

Biopsy specimens contain neutrophils in virtually all cases of H. pylori-positive cases if a sufficient number from both antrum and corpus is examined.

Neutrophils may be seen in the lamina propria, within the epithelium (particularly in the region of the glandular neck), and within the foveolar lumen, where they may form “pit abscesses.”

The density of intraepithelial neutrophils has been correlated with the extent of mucosal damage and with the intensity of H. pylori infection.

Neutrophils are a very sensitive indicator of the presence or absence of H. pylori and disappear within days of cure of infection.

If neutrophilic polymorphs are seen in a post-treatment biopsy but organisms are not apparent, a careful search for Helicobacter using one of the special stains or immunostains should be carried out.

Variable numbers of eosinophils infiltrate the lamina propria in most types of gastritis, but their pathogenetic role is unknown.

Therefore, routine grading of eosinophils is not required. Increased numbers of eosinophils are frequently seen in biopsy specimens from patients successfully treated for H. pylori gastritis; however, their density rarely approaches that found in eosinophilic gastritis (see below).

Chronic inflammation

The normal gastric mucosa contains only individual scattered chronic inflammatory cells (mononuclear cells) in the lamina propria. Thus, any increase indicates chronic gastritis.

In H. pylori infection, the cellular infiltrate contains effectors of the immune response, including CD4+ and CD8+ T-lymphocytes, B-lymphocytes, plasma cells, monocytes, mast cells, and eosinophils.

A few mononuclear leukocytes are always present in the lamina propria of the gastric mucosa; however, a precise definition of chronic inflammation is hampered by lack of a universal standard for the quantity of mononuclear inflammatory cells in the normal mucosa.

The latter can be heavily influenced by geographic location and other demographic variables of the persons studied and by observers’ subjective impressions.

From a pragmatic standpoint, it may be useful to think in terms of an “expected” rather than a “normal” level of chronic inflammatory cell infiltration.

Despite these limitations, the normal number of gastric mucosal mononuclear leukocytes in the lamina propria is viewed as a maximum of 2 to 5 lymphocytes, plasma cells and macrophages per highpower (×40 objective) microscopic field or, by another approach, two or three lymphocytes or plasma cells between foveolae (the area in which chronic inflammatory cells are most often found).

Plasma cells are sparse or absent from the stomach of healthy persons; so their presence is an especially important indicator of a chronic inflammatory response.

Some observers consider chronic inflammation to be present even when there are as few as one or two plasma cells per high-power field.

Occasional lymphocytes may also be observed in the epithelium of the normal stomach, especially in the surface (up to about 5 per 100 epithelial nuclei); when increased above that number, they also constitute chronic inflammation.

Grading the density of the chronic inflammatory cell infiltration provides a baseline measure with which to compare patient groups before treatment and with which to monitor the long-term effects of antibacterial therapy.

Chronic inflammatory cells have been shown to be slow to disappear after eradication of H. pylori, and may take a year or more to fall to “normal” levels.

Some researchers believe that even several years after cure the gastric mucosa, particularly in the antrum, remains infiltrated by a greater than normal number of chronic inflammatory cells.

The density of mononuclear cells in the lamina propria should be graded in areas away from lymphoid follicles and their surrounding marginal zone of small lymphocytes.

Glandular atrophy

Atrophy of the gastric mucosa is defined as loss of glandular tissue. Atrophy leads to thinning of the mucosa and is a common denominator in all pathological processes, causing severe mucosal damage.

Thus, the loss of glands may follow erosion or ulceration of the mucosa with destruction of the glandular layer or result from a prolonged inflammatory process where individual glands undergo destruction in a “piecemeal” fashion.

When this loss occurs, it may be followed by fibrous replacement or by a collapse of the existing supporting matrix.

Recognition of minor degrees of atrophy in the antrum is difficult because of the greater amount of connective tissue normally present in this compartment and the more irregular distribution of the pits and glands compared with the oxyntic mucosa.

A useful way to gauge antral atrophy is the demonstration that the three to four gland cross sections that normally span the lower antral mucosa are reduced to two or fewer cross sections.

Replacement of antral epithelium by intestinal metaplasia adds to the microscopic impression of atrophy and can be a useful indicator that atrophy is present, but the metaplasia itself is an independent process.

Separation of glands by chronic inflammation and fibrosis may make it difficult to recognize antral gland atrophy.

Atrophy in the oxyntic mucosa is closely linked to loss of acid secretion and to the development of intestinal metaplasia, which in turn is linked to an increased risk of gastric cancer.

Extensive atrophy in antral mucosa, usually associated with intestinal metaplasia, also carries an increased risk of malignancy.

Atrophy can be found in the absence of intestinal metaplasia, particularly in autoimmune gastritis, where there may be diffuse atrophy and pseudopyloric metaplasia in the corpus with minimal or no intestinal metaplasia. Atrophy should therefore be evaluated and graded independent of metaplastic changes.

Intestinal metaplasia

Intestinal metaplasia is common in chronic gastritis of all causes and increases in prevalence with disease duration.

Metaplastic epithelium can be recognized morphologically by the presence of goblet cells, absorptive cells, and cells resembling colonocytes or by its enzyme or mucin content.

Intestinal metaplasia has been categorized on the basis of morphology and enzyme histochemistry into small intestinal and colonic types or complete and incomplete forms and using mucin histochemistry into three main types according to its morphology and glycoprotein content .

Acidic glycoproteins in metaplastic cells are best demonstrated with the AB/PAS technique at pH 2.5 when they stain blue or purple, in contrast to the Schiff-positive neutral mucins present in the surface and foveolar epithelium and the mucous glands of the nonmetaplastic gastric mucosa.

In type I (which corresponds to complete, see glossary) goblet cells containing sialomucins are interspersed between nonsecretory absorptive cells with well-delineated brush borders.

In type II, sialomucin-containing goblet cells are scattered among gastric-type cells containing either neutral mucin or sialomucins;

type III is characterized by tortuous and branched crypts lined by tall columnar cells containing abundant sulfomucins with smaller numbers of goblet cells containing either sialomucins or sulfomucins.

Sulfomucins are differentiated from sialomucins by staining with high iron diamine (HID)/AB using the appropriate concentration of ferric chloride and carefully controlled conditions.

Intestinal metaplasia is generally regarded as a condition that predisposes to malignancy.

Complete (type I) metaplasia is believed to carry the lowest risk of gastric cancer, whereas those forms of metaplasia with large intestinal characteristics (type III metaplasia or incomplete forms) have been closely linked to carcinoma.

A follow-up study in Slovenia reveals that compared with individuals with types I and II sulfomucin-negative IM, subjects with type III IM have a 2.7 to 5.8 times greater risk for the development of gastric cancer.

Experience from other high-incidence countries, like Colombia and Japan, also point to a possible predictive value of type III metaplasia.

At this time, however, the performance of special stains to define the types of intestinal metaplasia and the assessment of gastric cancer risk should be confined to research settings.

More data are needed before specific guidelines for sampling strategy, endoscopic intervals, patient eligibility, and cost effectiveness of surveillance can be formulated.


A. The presence or absence of H. pylori, chronic inflammation, neutrophil polymorph activity, glandular atrophy, and intestinal metaplasia should be recorded in all cases of gastritis.

B. When present, each of these variables can be graded on a mild, moderate, or marked scale as indicated in the guidelines.

See also

- chronic gastritis