See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/261989318

The diagnosis of gastritis

Article in Diagnostic Histopathology · June 2014

DOI: 10.1016/j.mpdhp.2014.03.011

CITATIONS READS

0 455

3 authors, including:

Christina Falkeis
Klinikum Bayreuth GmbH
21 PUBLICATIONS 56 CITATIONS

SEE PROFILE

All content following this page was uploaded by Christina Falkeis on 24 February 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
 Author's personal copy
MINI-SYMPOSIUM: GASTRIC PATHOLOGY
The diagnosis of gastritis
Michael Vieth
Helmut Neumann
Christina Falkeis
Abstract
After Babylonian confusion over the histological classification of gastritis,
the Sydney system brought standardization and reproducibility to the
diagnostic field of gastric biopsies. Even some shortcomings do not
reduce the importance of the Sydney system for classification of gastritis.
Essentially gastritis is a purely histological diagnosis. Herein we describe
further diagnostic criteria for the diagnosis of gastritis and show the time-
dependent changes in frequencies of various types of gastritis over more
than 25 years at the Institute of Pathology in Bayreuth. Pathologists
should be encouraged to address the etiology of inflammatory infiltrates
to enhance the clinical value of a histological diagnosis on gastric
biopsies.
Keywords classification; etiology; gastritis; histology; Sydney system
Introduction
Gastritis is a histological diagnosis. Prerequisites are at least two
antral biopsies, each taken at 3 cm proximal the pyloric sphincter
from the lesser and greater curvature and two biopsies from the
corpus (body) close to the middle of the greater curvature.
Gastritis itself is a diagnosis that was used quite frequently in the
19th century but later it evolved into a clinical diagnosis to
describe upper abdominal complaints with widespread intro-
duction of endoscopes, biopsy techniques and the re-discovery of
Helicobacter pylori in 1983. The first description of gastric Heli-
cobacter dates back to Italy in 1892 by Bizzozero. In fact, in 1958
the Greek physician John Lykoudis introduced antibiotic treat-
ment for peptic ulcer disease and to “treat the infectious etiology
of gastritis”.1 The diagnosis of gastritis has thereby evolved into
confirming a possible etiology for the observed inflammation to
inform further therapy.
The updated Sydney system is the classical classification
system for grading and diagnosing gastritis. It was first intro-
duced in 19902,3 and revised in 1994.4 Before 1990, diagnosis of
gastritis was non-standardized, non-validated and often not
clinically relevant. The main problem was that diagnoses could
not be compared worldwide. In addition to the histological
component of the Sydney system, an endoscopic component was
introduced but this was quickly abandoned with the realization
that endoscopic criteria were poorly reducible and seldom
correlated with the histologically based etiology of the gastritis.5
Michael Vieth MD Director, Institute of Pathology, Bayreuth Clinic,
Bayreuth, Germany. Conflicts of interest: none declared.
Helmut Neumann MD Consultant, Internal Medicine I, Erlangen Univer-
sity, Erlangen, Germany. Conflicts of interest: none declared.
Christina Falkeis MD Consultant, Institute of Pathology, Klinikum
Bayreuth, Bayreuth, Germany. Conflicts of interest: none declared.
DIAGNOSTIC HISTOPATHOLOGY 20:6
The Sydney system brought convincing diagnostic tools and
descriptions in the form of a visual semi-quantitative scale for
pathologic evaluation. These allowed the pathologist standardi-
zation and validation of the diagnosis of gastritis.6
Before the Sydney system was introduced, for example, the
German ABC-scheme of gastritis gave at least the etiological base
for gastritis, such that A for stood for autoimmune gastritis, B for
bacterial (later: Helicobacter) gastritis and C for chemical reac-
tive gastritis.
It turned out that the ABC-scheme was the perfect addendum
to the Sydney system with its easily reproducible grading scale
that provided the etiology for the gastritis as an integral part of
the diagnosis to such that the clinicians could easily plan further
therapeutic steps.
On the other hand, the Sydney system also has a few short-
comings. The Sydney system was designed for antral and corpus
biopsies only. There is no grading proposal for cardiac mucosa,
although whether such an additional scheme would add any
relevant information is debatable.7 Other shortcomings are related
to the grading itself: e.g. grading atrophy in the antrum is
extremely difficult and in the corpus there has been debate con-
cerning whether loss of glands close to lymphatic aggregates
should be regarded and graded as atrophy or simply inflammation
pushing glands aside.8 Shortcomings of the etiology mainly
revolve around the distinction between normal mucosa and
chemical/reactive gastritis and the status after successful eradi-
cation of Helicobacter (so called ex-Helicobacter gastritis or post-
Helicobacter gastritis). Validated criteria for these distinctions are
not readily available. It is extremely difficult for pathologists to
differentiate between gastritis with focal atrophy due to present or
former Helicobacter infection and subsequent scar-forming e.g.
after mucosal breaks and cases with true autoimmune gastritis
with loss of parietal and chief cells leading to atrophy and a status
of so called pre-atrophic autoimmune gastritis. This situation is
further complicated by the notion that some patients, for unclear
reasons, present with autoimmune gastritis with remnant islands
of preserved oxyntic mucosa in the corpus. Further shortcomings
also remain in the risk evaluation for gastric carcinoma. However,
with the help of the Sydney system the so called corpus dominant
Helicobacter gastritis showing more inflammation in the corpus or
so called pangastritis with equal distribution of inflammation in
antrum and corpus has been associated with a higher risk (up to
34 fold) for development of gastric carcinoma compared to the
classical antrum predominant Helicobacter gastritis.9 On the other
hand, it is known that finding intestinal metaplasia and/or atro-
phy close to the angularis zone at the lesser curvature of the gastric
body is a risk factor for gastric carcinoma but this zone is not one
of the biopsy sites of the Sydney system.4 This observation has led
to several more intensified biopsy protocols and scoring systems
such as OLGIM and OLGA. These scoring systems are somewhat
helpful in identifying patients at higher risk for gastric carcinoma
but these are most applicable to patients with Helicobacter
gastritis and lead to unrealistically high cancer risk scores in pa-
tients with chemical/reactive gastritis and intestinal metaplasia.10
Endoscopic diagnosis of gastritis
Endoscopy is essential in the diagnostic evaluation of gastritis.
However, only histopathological analysis of biopsy specimens
213 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy

MINI-SYMPOSIUM: GASTRIC PATHOLOGY

can offer reliable diagnosis of gastritis and preneoplastic gastric atrophy nor intestinal metaplasia. Sparse lymphatic aggregates
conditions. Very recently, the European Society of Gastrointes- can be seen in some cases, especially in corpus mucosa.
tinal Endoscopy, the European Helicobacter Study Group, the
European Society of Pathology, and the Sociedade Portuguesa de Chemical/reactive gastritis
Endoscopia Digestiva have combined efforts to develop evidence-
In chemical/reactive gastritis, which by definition is limited to
based guidelines for the management of patients with precan-
the antral mucosa, lymphocytes and plasma cells within the
cerous conditions of the stomach. These guidelines state that
lamina propria are slightly increased leading to the Sydney
standard white-light endoscopy cannot accurately differentiate
grading: “slightly chronic, not active gastritis”. There is a pos-
between and diagnose preneoplastic gastric conditions. Various
sibility of overdiagnosing normal gastric mucosa as chemical
gastric endoscopic appearances have been proposed to be reli-
reactive gastritis. In order to avoid this, further criteria should be
able for in vivo diagnosis of gastritis. These include antral nod-
fulfilled, namely increasing number lamina propria ascending
ularity for Helicobacter pylori gastritis or absence of gastric rugal
smooth muscle fibers, apical fibrosis, capillary ectasia and
folds and presence of visible vessels in the gastric mucosa for
foveolar hyperplasia of the surface epithelium4,12 (Figure 1).
severe atrophic gastritis. Nevertheless, while the former has a
It should be noted, however that after successful Helicobacter
predictive value of >90%, it is only present in a minority of
eradication, both antrum and corpus show slightly to moderately
patients. The latter suffers from a low sensitivity of 48% and
increased numbers of lamina propria lymphocytes and plasma
14%, in the gastric corpus and antrum, respectively.11 Within
cells but no active inflammatory changes and frequently basal
recent years, advanced endoscopic imaging techniques have
remnants of lymphoid follicles and aggregates and often a slight
been implemented into daily routine clinical practice. These
tendency towards regenerative changes. In some individuals,
include dye-based and dye-free chromoendoscopy techniques,
these changes may persist after eradication therapy, whilst in
optical magnification endoscopy, and optical biopsy techniques,
others can evolve over time (especially in the antrum) into a
including confocal laser endomicroscopy and endocytoscopy.
histological picture similar to that of chemical reactive gastritis. A
Various studies have already suggested that dye-based chro-
minority seems to normalize over time.13 So called serum scars
moendoscopy, particularly when used with optical magnifica-
tend to normalize within 3 months. In some individuals IgG
tion, can detect intestinal metaplasia. The most commonly used
antibodies can persist for years.14
dye-agents include methylene-blue, indigo-carmine, or acetic
acid, which are mostly applied via standard spraying catheters.
Ex-Helicobacter gastritis (prior or post-Helicobacter gastritis)
Recent data have also indicated that dye-free chromoendoscopy,
using either optical chromoendoscopy (i.e. Narrow Band Imag- The diagnosis of so called ex-Helicobacter gastritis can easily be
ing; NBI) or virtual chromoendoscopy (i.e. FICE, i-scan) have a made if previous reports indicate a prior active Helicobacter
good sensitivity and specificity for diagnosis of gastric lesions. gastritis and previous eradication therapy or antibiotic treatment
Similar results have also been shown for optical biopsy tech- for other reasons. From these cases it is possible to extrapolate to
niques, allowing high-power magnification of the tissue thereby identify cases with ex-Helicobacter gastritis even without the
allowing analysis of cellular and subcellular features of the knowledge of prior eradication therapy or prior proven active
gastric epithelium. In this context, it has also been shown that Helicobacter gastritis. Such cases display lymphoid aggregates
optical biopsy techniques have the potential to diagnose H. pylori and follicles in combination with a slight or moderate chronic
and associated type-B gastritis.12 gastritis of antral and corpus mucosa.13
Despite the ongoing development of advanced endoscopic
imaging techniques, the conventional, physical biopsy is not to Helicobacter gastritis
be replaced. The most widely accepted classification and grading
Active Helicobacter infection or Helicobacter gastritis should be
system of gastritis is reflected in the updated Sydney system. The
diagnosed whenever active inflammatory infiltrates can be
system recommends two biopsies from the antrum (3 cm from
seen. In H. pylori gastritis these infiltrates should be diffuse
the pylorus, greater and lesser curvatures), one from the incisura
rather than focal, in contrast to cases with Helicobacter heil-
angularis, and two additional biopsies from the corpus (one from
mannii infection, in which inflammatory infiltrates are more
lesser curvature, 4 cm proximal to the incisura, and from the
focal. A band-like inflammatory infiltration of the upper half of
middle of the greater curvature). Although the additional value of
the mucosa is typical. Chronicity of the inflammation is mani-
biopsies from the incisura remains controversial, biopsy sam-
fest by infiltration of lymphocytes and plasma cells. According
pling and clear labeling in separate vials is of pivotal importance
to the Sydney system, activity is diagnosed from the number of
as atrophic gastritis and intestinal metaplasia are regularly
neutrophilic granulocytes. Sparse neutrophilic granulocytes
inhomogeneously distributed throughout the stomach.4
within the lamina propria are graded as slight activity, infil-
tration into the surface epithelium is graded as moderate ac-
How to diagnose histologically
tivity and presence of foveolar abscesses is graded as marked
Since using the Sydney system remains controversial for routine activity. Chronicity depends of the number of lymphocytes and
evaluation, the following discussion is more practical. Cancer plasma cells. Sparse lymphocytes and plasma cells can be seen
risk assessment systems like OLGA or OLGIM will not be further in normal gastric mucosa, can make it difficult to differentiate
discussed. normal versus slight chronic gastritis. In our practice, we re-
Gastric mucosa without pathological changes usually shows gard all infiltrates above sparse as slight chronic infiltration
few lymphocytes and plasma cells esp. in the antrum and no (space in between the inflammatory cells about five

DIAGNOSTIC HISTOPATHOLOGY 20:6 214 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy

MINI-SYMPOSIUM: GASTRIC PATHOLOGY

Figure 1 Endoscopic view of erosive gastritis (a) and chemical-reactive gastritis (b). Histological features of chemical gastritis: Foveolar hyperplasia,
sclerosis, ascending muscle fibers and ectatic capillaries.

lymphocytes), in cases with moderate infiltration 2e3 lym-
phocytes would fit in between the inflammatory cells. In cases
with dense lymphocytic infiltration, in which only single or no
lymphocytes would fit in between the inflammatory cells,
grading is considered as marked.4 Learning curves apply. The
semi-quantitative analog scale of the updated Sydney system
can be useful in grading the inflammatory infiltrate and can be
made available as a ready reference next to the microscope15
(Figure 2). Special forms of active gastritis include infection
with Helicobacter heilmannii, which often leads to a rather mild
and focally active gastritis (Figure 3). Focal activity in the
gastric mucosa can also be due to self-limiting infectious non-

Figure 2 The updated Sydney system classifies the chronic inflammation (aec) as a ¼ slight, b ¼ moderate, c ¼ marked. The active inflammation (def ) is
classified as d ¼ slight, e ¼ moderate, f ¼ marked.

DIAGNOSTIC HISTOPATHOLOGY 20:6 215 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy

MINI-SYMPOSIUM: GASTRIC PATHOLOGY

Figure 3 Helicobacter pylori is typically located on the cell surface as displayed in picture (a), in contrast to Helicobacter heilmannii, which lies in the
middle of the pits (b).

Helicobacter gastritis. The infectious agents remain unknown. Autoimmune gastritis

Interestingly there are accumulations in spring and autumn and
Autoimmune gastritis should not be confused with focal atrophic
during PPI therapy. No specific symptoms or any further con-
gastritis due to scarring or pushed aside glands due to lymphoid
sequences are known. Thus, such a diagnosis can be regarded
aggregates or follicles.
as a rather harmless incidental finding.
Autoimmune gastritis should be diagnosed whenever complete
It must be noted that Crohn’s gastritis can present a similar
parietal and chief cell atrophy destroys the body mucosa. Some-
picture but, in this case, the chronic component is usually more
times it is difficult to differentiate antral mucosa from atrophic
pronounced than the active component compared to infectious
corpus mucosa. Therefore, when the suspicion for autoimmune
non-Helicobacter gastritis. Diagnosis of gastric Crohn’s disease
gastritis is high, the gastroenterologist may be asked for fundic and
should be made only when status of the lower GI-tract is known.
corpus biopsies to be sent in different vials to allow differentiation
If the lower tract status in unknown, to the findings should be
between antrum and corpus mucosa or, alternatively, a gastrin
described and a notation should be made to perform lower GI-
stain may be added (G cells are absent in the gastric body and
Tract endoscopy and exclude Crohn’s disease.16 A learning
present in the antrum so only antral samples display positive cells
curve applies for classification of the various forms of gastritis. At
with a gastrin stain). In many cases intestinal metaplasia can be
our institution we have learned that most residents need to re-
seen as well17 (Figures 4 and 5). In autoimmune gastritis, besides
view at least 1500 gastric biopsies to become acquainted with the
atrophy, one should expect more or less dense lymphocytic in-
various patterns of gastritis and to acquire confidence in applying
filtrates in the lamina propria, foveolar hyperplasia and the
the Sydney system.

Figure 4 Endoscopic view of atrophic gastritis (a) and intestinal metaplasia (b).

DIAGNOSTIC HISTOPATHOLOGY 20:6 216 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy
MINI-SYMPOSIUM: GASTRIC PATHOLOGY
presence of small proliferating neuroendocrine cell complexes.
Immunohistochemistry is not necessary to detect these prolifer-
ating small complexes. If a neuroendocrine tumor is present,
immunohistochemistry for ki67 is recommended to assess the
proliferation index and mitoses should be counted as an integral
part of information for the TNM classification.18 Proof of the
neuroendocrine nature of a lesion can be determined by immu-
nohistochemistry, although it is usually not necessary since le-
sions are very characteristic on routine H&E stains. Exceptions
apply to cases for which it may be necessary to analyze the hor-
mones that are secreted by the lesion. Diagnosis of autoimmune
gastritis can become rather difficult when some parts of the body
mucosa show preservation of the parietal cells. There are two
forms of partially or fully preserved body (oxyntic) glands. One
form consists of islands of fully preserved corpus mucosa with
characteristic hypertrophy of parietal cells. These islands have a
polypoid appearance on endoscopy and can easily be mistaken for
neuroendocrine tumors or other types of polyps endoscopically.
These islands are more easily identified when the information is
given that the individual is not on PPI medication and has atrophic
gastric mucosa. If in doubt in these cases, separate corpus and
fundic biopsies may help to diagnose the underlying autoimmune
gastritis. The second form is discussed more controversially: pre-
atrophic autoimmune gastritis with dense lymphocytic infiltra-
tion and continuous slight or moderate atrophy of the body mu-
cosa. Various degrees of active inflammation are possible.
Intestinal metaplasia can be seen sometimes. Helicobacter can be
present or absent. The antrum shows a slight chronic gastritis with
remnants of basal lymphoid aggregates or follicles and thus diag-
nosis of ex-Helicobacter-gastritis can be made in antrum. These
pre-atrophic cases can often be cured by eradication therapy even if
Helicobacter cannot be detected morphologically. Serology against
Helicobacter antibodies may help in such a situation. If Heli-
cobacter antibodies are present, eradication therapy is even more
justified. We have seen complete cure of gastritis and atrophy after
6e9 months after successful eradication therapy.17
Special forms of gastritis
Lymphocytic gastritis: lymphocytic gastritis, with more than 20
lymphocytes per 100 epithelial cells, is significantly associated
Figure 5 Pre-atrophic autoimmune gastritis with remnants of corpus mucosa (a). Complete intestinal metaplasia in autoimmune gastritis with goblet cells
and Paneth cells at the bottom (b). Typical changes in autoimmune gastritis (c): pancreatic-acinar metaplasia (arrows) and hyperplasia of neuroendocrine
cells (star).
DIAGNOSTIC HISTOPATHOLOGY 20:6
with celiac disease. The classical form of gastritis varioliformis
with small protrusions and central erosions is a rare finding. The
diagnosis can already be suspected endoscopically. The vast
majority of patients whose biopsies show lymphocytic gastritis
present endoscopically with either normal mucosa, reddened
mucosa or even mucosa with giant folds.4 Duodenal biopsies
should always be performed to exclude celiac disease. Heli-
cobacter can be present or absent. In cases with celiac disease,
lymphocytic gastritis normalizes on a gluten-free diet if the pa-
tients lack Helicobacter infection. In patients without celiac dis-
ease, eradication therapy helps to normalize the epithelium
within a short time after eradiation even if Helicobacter was not
detected morphologically.19 The vast majority of cases of auto-
immune gastritis are believed to be Helicobacter induced. Cross-
reacting antibodies against parietal and chief cells induce atrophy
of the gastric body. Interestingly, patients with autoimmune
gastritis present with iron deficiency anemia long before vitamin
B12 deficient (megaloblastic) anemia can be detected.20 Early on,
the gastric body is still fully preserved but shows Helicobacter
infection only. Rare individuals have congenital absence of
gastric body type mucosa. Most patients develop atrophy over
time (often decades). In patients who present with full loss of
their parietal cell mass, eradication therapy will not improve
their mucosa. Eradication therapy has been proven to have an
effect only when there is remnant body mucosa.18
Infectious non-Helicobacter gastritis: this form is more often
seen in spring and autumn. The infectious agent has yet to be
identified. Normally, this form is self-limiting, vanishing after 2
weeks and is a truly incidental finding. After PPI therapy, such
cases can be observed more frequently. Specific symptoms are
not known. Characteristically focal active inflammation, often
with foveolar abscesses, is detected.
Collagenous gastritis: this is a very rare form of gastritis pre-
senting with a continuous or discontinuous thickened sub-
epithelial collagenous band. This form can be combined with
collagenous duodenitis and colitis. Specific symptoms are not
known nor any form of cure for collagenous gastritis. Collage-
nous gastritis is probably caused by an infectious agent that has
not yet been identified.21
217 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy
MINI-SYMPOSIUM: GASTRIC PATHOLOGY
Crohn’s gastritis: this diagnosis should not be made in the
absence of information on the lower GI-tract. Without such
knowledge, only a suspicion for Crohn’s gastritis should be
offered. The differential diagnosis includes infectious non-
Helicobacter gastritis. Usually one would expect a more active
inflammation in infectious non-Helicobacter gastritis than in
Crohn’s disease. Crohn’s gastritis is focal (predominantly peri-
glandular) also but shows more predominant chronic inflam-
mation with lymphocytes and plasma cells (Figure 6). The
problem with Crohn’s gastritis is that it shows a huge variation.
Cases with extensive inflammation should not be mistaken for
Helicobacter gastritis. Careful search for Helicobacter should be
undertaken in cases suspicious for Crohn’s gastritis. If active
Helicobacter infection is detected the diagnosis of Crohn’s
gastritis is not be possible in most cases. In this situation, erad-
ication therapy is recommended to clarify Crohn’s involvement
of the upper GI-tract. If in doubt, biopsies from the upper GI-tract
(antrum/corpus/duodenum) may help to clarify the diagnosis of
Crohn’s disease since 70% of patients with Crohn’s disease are
believed to show some involvement of the upper GI-tract. It
should be noted that giant cells or granulomas are only rarely
found in the upper GI-tract.16
Clinical significance of gastritis diagnosis
It must be noted that the clinical significance of a gastritis diag-
nosis is rather limited. Besides exclusion of neoplasia, clinicians
are mostly interested in confirming or excluding active Heli-
cobacter infection and autoimmune gastritis. These two di-
agnoses can have either therapeutic impact in the form of an
eradication therapy or, in the case of autoimmune gastritis, some
follow-up strategies. All other diagnoses imply mostly no further
therapeutic steps.22,23 Furthermore it must be accepted that cli-
nicians are not interested in grading of gastritis but in the
Figure 6 Granuloma in Crohn’s gastritis (a) and typical basal inflammation of the pits (b).
DIAGNOSTIC HISTOPATHOLOGY 20:6
underlying etiology of the inflammation. Therefore, a clear eti-
ology of the observed inflammatory infiltrates should be offered
whenever possible. A descriptive diagnosis like: “slight chronic
not active gastritis, Helicobacter negative” does not make any
clinical sense. It should be noted that antral and corpus biopsies
do deserve separate diagnoses. It must be in the primary interest
of the pathologist to give the clinician as much information as
possible on the etiology of inflammatory infiltrates to enhance
the significance of a histological gastritis diagnosis. Special in-
terest should be paid to the special forms of gastritis that are not
included in the Sydney system. It should be kept in mind that
there is no grading scheme for cardia biopsies. Regardless, even
in the cardia it is sometimes possible to diagnose inflammatory
infiltrates characteristic for Helicobacter infection. In these cases,
of course the etiology should be reported. More difficult and still
somewhat unclear are gastric cardiac inflammatory infiltrates
associated with reflux disease. Data in the literature are very
confusing and even contradictory. Thus, other than Helicobacter
infection, an etiology in cardiac gastritis should not be reported.7
Development of gastritis with time
The Institute of Pathology in Bayreuth serves as a tertiary reference
center for gastrointestinal pathology, especially for gastritis since
1967. Kurt Elster, who was the first director of the institute, was
also the first pathologist to examine gastric biopsies and create
diagnostic schemes for gastric neoplasia and gastritis. The second
director, Manfred Stolte, was involved directly in the Helicobacter
story and in developing classification schemes such as the Sydney
system. Michael Vieth, as the third director, works mainly on the
inflammation at the cardia and subsequent distal esophagus to
round up the scientific interests in the stomach in Bayreuth.
The re-discovery of Helicobacter in 1983 and the increasing
knowledge of the pathogenic potential of Helicobacter led to
218 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy

MINI-SYMPOSIUM: GASTRIC PATHOLOGY

Number of cases with gastric biopsies per year (2013a: through 31.7.2013)
2013a 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001

41,843 65,625 66,356 63,683 62,651 73,798 75,493 70,928 66,327 61,406 61,586 61,811 61303
2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 1990 1989 1988 1987
60,814 56,851 54,751 51,838 52,452 51,055 44,552 36,366 30,626 23,706 19,224 14,033 3336 1421
a
Data on 2013 are available only till 31. July of 2013 but not for the whole year!.

Table 1

antibiotic eradication schemes very soon afterwards. Indications
for eradications are clearly given in national and international
guidelines (such as for children with Helicobacter infection,
peptic ulcer disease, prior neoplasm, prior gastric operation, pre-
atrophic gastritis and suspicion for Crohn’s gastritis, among
others). Discrepancies are mostly due to the rate of infection in
various populations and also to the frequency of presence of
corpus-predominant gastritis with more marked inflammatory
infiltrates in corpus than in antrum. Patients with this special
form of Helicobacter gastritis have the highest risk of developing
malignancy compared to all other indicators for progression to-
wards malignancy and distribution of inflammation may explain
why there are high risk populations in some areas depending on
patients’ genetic background and on the Helicobacter strain that
leads to corpus-predominant gastritis.9,22
With introduction of widely available eradication schemes
one should expect that the diagnosis of ex-Helicobacter-gastritis
should become more frequent with time. Taking into account
that most cases of autoimmune gastritis are Helicobacter induced
one would expect a lower rate of autoimmune gastritis than some
time ago.
In an attempt to clarify changes in the diagnosis of gastritis, we
searched diagnoses of gastritis from the archives of the Institute in
Bayreuth since standardized diagnoses have been made. We
analyzed gastritis diagnoses from 1987 through mid 2013,
including more than 1.3 million individuals (see Table 1) with
gastric biopsies. Interestingly, not all of the above assumptions
were confirmed (Figure 7). It is obvious that the diagnosis of ex-
Helicobacter-gastritis increased over time. However, the fre-
quency of autoimmune gastritis has remained stable over the last
decades. There has been no improvement despite widely applied
eradication therapies. A possible explanation could be that more
time is needed to observe an effect of eradication therapies since
the development of autoimmune gastritis also requires decades.
Alternatively, some cases of autoimmune gastritis may result from
an alternate trigger that Helicobacter. Another interesting finding
is that, to date, cases of ex-Helicobacter gastritis have not sur-
passed the number of cases with active Helicobacter infection. In
2006 it appeared that the ex-Helicobacter-gastritis curve would
meet the Helicobacter curve but in the last few years the infection
rate accounts for about 50% (see Figure 6 and Table 1). Several
factors may explain this. First, not all detected infections are
treated, secondly, immigrants from countries with higher preva-
lence may dilute results and thirdly the increasing numbers of
multi-resistant infections hamper the success of eradication ther-
apies in a way that there are still more active Helicobacter

40.000

30.000 A-Gastritis
Number of Diagnoses

NORMAL

20.000 Ex-Hp-Gastritis

Hp-Gastritis

C-Gastritis
10.000
Lymphocytic
Gastritis

0
1987 1991 1995 1999 2003 2007 2011
1989 1993 1997 2001 2005 2009 2013
year

Figure 7 Development of gastritis over time (n ¼ 1,333,839 patients). Lines are given as absolute numbers of diagnoses per year (A-gastritis ¼ auto-
immune gastritis; normal ¼ stomach with no pathological changes; Ex-Hp-Gastritis ¼ gastritis after successful eradication; C-Gastritis ¼ chemical reactive
gastritis).

DIAGNOSTIC HISTOPATHOLOGY 20:6 219 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy

MINI-SYMPOSIUM: GASTRIC PATHOLOGY

Number of cases with gastric carcinomas per year (2013a: through 31.7.2013) subdivided into cases with and without
autoimmune gastritis. Total number of gastric carcinomas 1987e2013: 7687 cases
Year 2013a 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001

A-gastritis 5 9 15 14 9 14 21 23 12 8 15 9 6
All cases 151 201 248 239 212 322 408 409 412 351 366 399 392
Year 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 1990 1989 1988 1987
A-gastritis 8 3 1 0 3 2 11 10 6 11 11 9 3 0
All cases 381 415 261 138 178 187 247 205 212 222 304 237 270 320
a
Data on 2013 are available only till 31. July of 2013 but not for the whole year!.

Table 2

infections than cases with ex-Helicobacter-gastritis. It has to be Gastric carcinoma

noted that these data are from a German population and compared
to the US or the UK, the situation is very much different. In the US
only a subfraction of the population is still Helicobacter infected
such that the incidence of gastric carcinoma is diminishing.24
Therefore, in the US, gastric biopsies to assess for Helicobacter
infection are not cost effective. Europe, however, remains in a
different position concerning the rate of infection. Asia is the other
extreme. There, one can observe high risk populations for which
screening is justified to eradicate and cure gastric neoplasia at its
earliest stage. In other areas of the world, screening is not recom-
mended either due to low risk rate of infection or low risk for
malignant progression (such in Thailand with high rates of in-
fections but low risk for malignant transformation).25 These ob-
servations go in parallel with the rate of corpus-predominant
gastritis, with an odds ratio of 34 for carcinoma. Focal atrophy and
intestinal metaplasia increase the odds ratio for carcinoma by
about 5e6 fold and are thus are rather weak indicators for a pre-
malignant status. Nevertheless it must be noted that focal atrophy
and intestinal metaplasia are bystanders after marked mucosal
damage and thus serve as indirect risk markers e.g. in OLGA or
OLGIM to obtain a better risk stratification for patients with
ongoing gastric inflammation to decide on further therapeutic
steps such as eradication therapy or intensified follow-up schemes
according to national and international guidelines.22,23
More difficult to interpret is the curve of frequency of chem-
ical reactive gastritis. It seems that there is a considerable
amount of overdiagnosis of chemical reactive gastritis in Bayr-
euth on the basis of minor histological changes. Criteria for
diagnosis of chemical/reactive gastritis have been described in
the literature4,26 but it seems that in our institution we are rather
quick to make such a diagnosis even without the clinical
knowledge of NSAID/ASA intake. The rather low number of
“normals” without any pathological changes may support this
notion of possible overdiagnosis. Fortunately the diagnosis of
chemical/reactive gastritis has no real clinical meaning since
patients with chemical reactive gastritis or without any patho-
logical changes receive no follow-up nor any further therapeutic
steps. However, our data clearly demonstrate that diagnoses of
chemical reactive gastritis, ex-Hp-gastritis and cases without
pathological changes may show considerable overlapping due to
non-existing or non-described or non-validated criteria. Further
studies are needed to grade those minute changes and ascribe the
correct etiology. This would be of basic research interest, albeit
with virtually no clinical impact.
Between 1987 and Sept. 2013, 19,940 cases of gastric carcinoma
were diagnosed at our institution.
Interestingly, we observed a peak of the number of gastric
carcinomas during the years 1999e2007 (Table 2) with steady
decrease since then. Parallel to this increase, the proportion of
gastric carcinoma in patients with autoimmune gastritis showed
the same increase and decrease. The decrease in the last years can
be explained by the effect of Helicobacter eradication therapy. The
increase from 1999 to 2007 is explained by the fact that the number
of endoscopies rose markedly since the mid 1990s (Table 1).
Conclusion
In summary, histological description of gastric inflammation has
almost no real clinical value. It is the etiology of the infiltrates
that matters and impacts treatment. In particular, diagnosing the
special forms of gastritis may increase the reputation of pathol-
ogists since these diagnoses are sometimes difficult to make but
can have larger clinical impact and consequences. Interestingly,
even after clarifying that most cases with autoimmune gastritis
are Helicobacter induced, eradications therapies did not signifi-
cantly alter the frequency of autoimmune gastritis. May be this
will require more time to be observed than anticipated. Regard-
less, in Germany and probably in Europe due to its relatively
homogenous population, active Helicobacter infection is still the
most frequent form of gastric inflammation.22 For cardiac bi-
opsies, the etiology cannot typically be given despite clear Heli-
cobacter sequel.7 At least two antrum and corpus biopsies each
need to be taken to ensure a correct gastritis diagnosis. With
these four biopsies as a minimum a gastritis diagnosis will be
correct around 90e95% of the time. When fewer than these four
biopsies are obtained and biopsies from the antrum or corpus are
missing, the etiology and diagnosis may be incorrect.4 It is one of
the primary tasks of the pathologist to spread this message to
clinical colleagues and insist on the number and locations of
biopsies to ensure the clinical utility of a histological gastritis
diagnosis. A
REFERENCES
1 Marshall B. Helicobacter pioneers. Victoria, Australia: Blackwell Sci-
ene Asia Pty Ltd, 2002; 1e203.
2 Price AB. The Sydney system: histological division. J Gastroenterol
Hepatol 1991; 6: 209e22.

DIAGNOSTIC HISTOPATHOLOGY 20:6 220 Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy
MINI-SYMPOSIUM: GASTRIC PATHOLOGY
3 Tytgat GNJ. The Sydney system: endoscopic division, endoscopic
appearance in gastritis/duodenitis. J Gastroenterol Hepatol 1991; 6:
223e34.
4 Dixon MF, Genta RM, Yardley JH, Correa P, and the Participants in the
international Workshop on the Histopathology of gastritis, classifi-
cation and grading of gastritis. The updated Sidney system. Am J
Surg Pathol 1996; 27: 1161e81.
5 Khakoo SI, Lobo AJ, Sheperd NA, Wilkinson SP. Histological assessment
of the Sydney classification of endoscopic gastritis. Gut 1994; 1172e5.
6 Lash JG, Genta RM. Adherence to the Sydney system guidelines in-
creases the detection of Helicobacter gastritis and intestinal meta-
plasia in 400738 sets of gastric biopsies. Aliment Pharmacol Ther
2013; 38: 424e31.
7 Peitz U, Vieth M, Malfertheiner P. Carditis at the interface between
GERD and Helicobacter pylori infection. Dig Dis 2004; 22: 120e5.
8 Rugge M, Correa P, Dixon MF, et al. Gastric mucosal atrophy: inter-
observer consistency using new criteria for classification and grading.
Alimentary Pharmacol Ther 2002; 16: 1249e59.
9 Imagawa S, Yoshihara M, Ito M, et al. Evaluation of gastric cancer risk
using topography of histological gastritis: a large-scaled cross-
sectional study. Dig Dis Sci 2008; 53: 1818e23.
10 Rugge M, Fassan M, Pizzi M, et al. Operative link for gastritis
assessment vs operative link on intestinal metaplasia assessment.
World J Gastroenterol 2011 Nov 7; 17: 4596e601.
11 Dinis-Ribeiro M, Areia M, de Vries AC, et al. Management of pre-
cancerous conditions and lesions in the stomach (MAPS): guideline
from the European Society of Gastrointestinal Endoscopy (ESGE),
European Helicobacter Study Group (EHSG), European Society of
Pathology (ESP), and the Sociedade Portuguesa de Endoscopia
Digestiva (SPED). Endoscopy 2012; 44: 74e94.
12 Urquhart P, DaCosta R, Marcon N. Endoscopic mucosal imaging of
gastrointestinal neoplasia in 2013. Curr Gastroenterol Rep 2013;
15: 330.
13 Kodama M, Murakami K, Okimoto T, et al. Ten-year prospective
follow-up of histological changes at five points on the gastric mucosa
as recommended by the updated Sydney system after Helicobacter
pylori eradication. J Gastroenterol 2012; 47: 394e403.
14 Koizumi W, Tanabe S, Imaizumi H, et al. Effect of anti-Helicobacter
pylori IgG antibody titer following eradication of Helicobacter pylori
infection. Hepatogastroenterology 2003; 50: 293e6.
DIAGNOSTIC HISTOPATHOLOGY 20:6
View publication stats
15 el-Zimaity HM, Graham DY, al-Assi MT, et al. Interobserver variation in
the histopathological assessment of Helicobacter pylori gastritis.
Hum Pathol 1996; 27: 35e41.
16 Danelius M, Ost A, Lapidus AB. Inflammatory bowel disease-related
lesions in the duodenal and gastric mucosa. Scand J Gastroenterol
2009; 44: 441e5.
17 Stolte M, Baumann K, Bethke B, Ritter M, Lauer E, Eidt H. Active
autoimmune gastritis without total atrophy of the glands.
Z Gastroenterol 1992; 30: 729e35.
18 Hosokawa O, Kaizaki Y, Hattori M, et al. Long-term follow up of pa-
tients with multiple gastric carcinoids associated with type A
gastritis. Gastric Cancer 2005; 8: 42e6.
19 Hayat M, Arora DS, Dixon MF, Clark B, O’Mahony S. Effects of Heli-
cobacter pylori eradication on the natural history of lymphocytic
gastritis. Gut 1999; 45: 495e8.
20 Hershko C, Ronson A, Souroujon M, Maschler I, Heyd J, Patz J. Vari-
able hematologic presentation of autoimmune gastritis: age related
progression from iron deficiency to cobalamin depletion. Blood
2006; 107: 1673e9.
21 Leung ST, Chandan VS, Murray JA, Wu TT. Collagenous gastritis:
histopathologic features and association with other gastrointestinal
diseases. Am J Surg Pathol 2009; 33: 788e98.
22 Moehler M, Al-Batran SE, Andus T, et al. German S3-guideline
“Diagnosis and treatment of esophagogastric cancer”.
Z Gastroenterol 2011; 49: 461e531.
23 Malfertheiner P, Megraud F, O’Morain CA, et al. European Heli-
cobacter Study Group. Management of Helicobacter pylori infec-
tionethe Maastricht IV/Florence Consensus report. Gut 2012; 61:
646e64.
24 Bashinskaya B, Nahed BV, Redjal N, Kahle KT, Walcott BP. Trends
in peptic ulcer disease and the Identification of Helicobacter
pylori as a causative organism: population-based estimates from
the US Nationwide Inpatient sample. J Glob Infect Dis 2011 Oct;
3: 366e70.
25 Hirai I, Sasaki T, Kimoto A, et al. Infection of less virulent Helicobacter
pylori strains in asymptomatic healthy individuals in Thailand as a
potential contributing factor to the Asian enigma. Microbes Infect
2010; 12: 227e30.
26 Borchard F. Chemisch reaktive gastritis. Pathologe 2001; 22: 44e55.
221 Ó 2014 Elsevier Ltd. All rights reserved.