Dig Dis Sci (2009) 54:19581965
DOI 10.1007/s10620-008-0553-y
ORIGINAL ARTICLE
In French Children, Primary Gastritis Is More Frequent
Than Helicobacter pylori Gastritis
N. Kalach S. Papadopoulos E. Asmar C. Spyckerelle
P. Gosset J. Raymond E. Dehecq A. Decoster
C. Creusy C. Dupont
Received: 14 May 2008 / Accepted: 16 September 2008 / Published online: 12 November 2008
Springer Science+Business Media, LLC 2008
Abstract The aim of this study was to analyze the his-
tological characteristics according to the updated Sydney
classification (intensity of gastritis, degree of activity,
gastric atrophy, intestinal metaplasia, and Helicobacter
pylori) in symptomatic children referred for upper gastro-
intestinal endoscopy. A 4-year retrospective descriptive
study was carried out in 619 children (282 females and 337
males), median age 3.75 years (15 days to 17.3 years)
referred for endoscopy. Six gastric biopsies were done
(three antrum and three corpus) for histological analysis
(n = 4), direct examination and H. pylori culture (n = 2).
H. pylori status was considered positive if at least two out
of three tests were positive and negative if all three tests
were negative. The results showed that only 66 children
N. Kalach (&)
E. Asmar
C. Spyckerelle
Department of Pediatrics, Saint Antoine Paediatric Clinic,
Saint Vincent de Paul Hospital, Catholic University of Lille,
P.O. Box 387, Bd de Belfort, 59020 Lille Cedex, France
e-mail: kalach.nicolas@ghicl.net
N. Kalach
C. Dupont
Department of Pediatrics, Neonatalogy Cochin-Saint Vincent de
Paul Hospital, University Paris Descartes, Paris, France
S. Papadopoulos
P. Gosset
C. Creusy
Department of Pathology, Saint Vincent de Paul Hospital,
Catholic University of Lille, Lille Cedex, France
J. Raymond
Department of Microbiology, Cochin-Saint Vincent de Paul
Hospital, University Paris Descartes, Paris, France
E. Dehecq
A. Decoster
Department of Microbiology, Saint Philibert Hospital,
Catholic University of Lille, Lille Cedex, France
123
(10.66%) were H. pylori positive. Histological antral and
corpus gastritis was detected in, respectively, 53.95% and
59.12% of all cases, most of them of mild grade 1. Antral
and corpus activity was grade 1 in 18.57% and 20.03% of
cases. H. pylori-positive versus H. pylori-negative children
did differ in terms of moderate and marked histological
gastritis and grade 2 or 3 activities. One girl had moderate
gastric atrophy and another one moderate intestinal meta-
plasia, both being H. pylori negative. The findings indicate
that primary antrum and corpus gastritis is 5.3 and 6.9
times, respectively, more frequent than H. pylori gastritis
in French children, with usually mild histological gastritis
and activity. Gastric atrophy and intestinal metaplasia are
rare.
Keywords Helicobacter pylori
Children
Gastritis

Activity
Primary
Gastric atrophy
Intestinal metaplasia

Sydney classification
Introduction
The updated Sydney classification is now well established
as the most appropriate for the precise description of his-
tological gastric biopsy specimens [1]. In children,
prevalence of histological chronic gastric inflammation
varies from 75% to 100% [25]. The majority of authors of
pediatric studies reported histological gastritis in all cases
of Helicobacter pylori infection [3, 68]. However, some
cases of infection without microscopic lesions were also
reported [4, 9]. H. pylori infection is found in most prim-
itive gastritis in adults, whereas it was observed in only
60% of pediatric gastritis [7, 9]. This difference, potentially
linked to lower infection prevalence in children, also
reflects the probable existence of other causes of gastritis,
Dig Dis Sci (2009) 54:19581965
which could be masked in adults by the strong incidence of
subjects carrying H. pylori. The detection of primary antral
gastritis (PAG) in about 20% of children with recurrent
abdominal pain (RAP) [10], and the discovery of other
types of germs in the mucous membrane of some subjects
with gastritis, could support this assumption [11]. The low
concentration of germs on the surface of the mucous
membrane related to the recent character of the infection in
the child could also explain the high histological rate of
gastric inflammation without H. pylori isolation [9].
Mucosal atrophy and intestinal metaplasia of the stom-
ach are frequently found in gastric biopsy specimens of
adults infected with H. pylori [12]. However, if H. pylori
infection is frequently acquired during childhood, its role in
gastric atrophy and intestinal metaplasia has not been
extensively searched for and described [13].
The purpose of this study was to analyze the charac-
teristics and parallelism of clinical, endoscopic, and
histological features according to the updated Sydney
classification [1] in symptomatic children referred to a
pediatric gastroenterology clinic with clinical symptoms
leading to upper gastrointestinal (GI) endoscopy and
biopsy.
Patients and Methods
Patients
A retrospective descriptive study was carried out in the
pediatric department of Saint Vincent de Paul Hospital,
Lille, France. This study enrolled all children: (1) referred
for upper GI endoscopy between January 2001 and
December 2004 in the course of diagnostic and etiologic
assessment of clinical gastritis, i.e., the association of one
or more symptom or sign of those digestive manifestations:
RAP (at least three episodes of abdominal pain per week
during the last 3 months), gastro-oesophageal reflux
(GER), vomiting, occult or macroscopic digestive hemor-
rhages, failure to thrive, or other digestive demonstrations
of gastritis, anorexia, malaise, dysphasia; (2) whose
endoscopy included gastric biopsy specimens. Children
were excluded from analysis if they had received antibi-
otics, anti-H2, or proton pump inhibitors (PPIs),
nonsteroidal anti-inflammatory drugs (NSAID) 4 weeks
before endoscopy or if there was presence of chronic
organic diseases, i.e., cystic fibrosis, metabolic diseases,
inflammatory bowel diseases, celiac disease, diabetes
mellitus, etc. Ethnic origin of children was identified by
mothers country of birth, i.e., Caucasian for Europe, or
non-Caucasian for Middle East or Africa. In accordance
with good clinical practice, written consent was obtained
from all parents before endoscopy.
1959
Endoscopy, Gastric Biopsy, and Histological Analysis
Endoscopic lesions were recorded as follows: for esopha-
gus, normal, inflammatory esophagitis (grade I), ulcerative
esophagitis (grade II), mycotic esophagitis, endobrachy-
esophagus, and esophageal varices; for gastric lesions,
normal, gastritis (congestive, nodular, and erosive) and
ulcer; for bulbar lesions, normal, bulbitis (congestive,
nodular, atrophic), and bulbar ulcer; for duodenal lesions,
normal, duodenitis (congestive, nodular, atrophic), and
duodenal ulcer.
Biopsies were performed in both gastric antrum (n = 3)
and corpus (n = 3). Two antral and two corpus specimens
were formalin fixed and paraffin embedded, and hema-
toxylin and eosin stains were used for histopathologic
analysis according to updated Sydney classification [1].
The amounts of inflammation, activity, atrophy, meta-
plasia, and H. pylori were established by the visual analog
scale given in the updated Sydney classification [1].
Inflammation
Intensity of histological gastric inflammation, i.e., histo-
logical gastritis, was estimated according to mononuclear
cells and lymphocytes, i.e., absent (grade 0), few (1, mild),
moderate number (2, moderate), or high number (3,
marked).
Activity
Degree of activity, i.e., histological active gastritis, was
estimated according to neutrophils infiltration, i.e., absent
(grade 0), few (1, mild), moderate number (2, moderate), or
high number (3, marked).
Gastric Glandular Atrophy and Intestinal Metaplasia
Gastric glandular atrophy and intestinal metaplasia were
assessed as absent (grade 0), mild (1), moderate (2), or
marked (3).
Helicobacter pylori
Intensity of infection with H. pylori was classified as
Gram-negative bacillus absent (grade 0), mild or rare (1),
moderate (2), or marked (3).
Further histological analysis was done by recording
presence or absence of follicular gastritis.
H. pylori Status: Direct Examination and Culture
One antral and one corpus gastric biopsy specimens were
placed in a tube of sterile physiological saline solution,
123
1960 Dig Dis Sci (2009) 54:19581965

transported to the laboratory of bacteriology, maintained at

4C, then quickly dealt with. The bacteriological exami-
nation included direct examination of the totality of a first
biopsy specimen with the use of Gram staining at 1% in a
concentrate of carbolfuschin (RAI, Paris, France). The
biopsies were put thereafter in culture in 0.5 ml meat-liver
suspension (Pasteur, Paris, France), homogenized by
Griffith tube and cultivated on chocolate gelose (Bio-
Merieux, Paris, France) then incubated with 37C in
microaerophilic conditions (5% O2, 10% CO2, 85% N2)
during 7 days. The presence of small, round, convex col-
onies with strong urease activity, oxidasic, catalasic,
gamma glutamyl transferasic, and alkaline phosphatasic
with a negative nitrate reductase reaction (API, Bio-Mer-
ieux, Paris, France) enabled confirmation of diagnosis of
H. pylori infection.
The results of the histological analysis, bacterial culture,
and direct examination of the gastric biopsy specimens
Fig. 1 Distribution of the age scales and antral and corpus histologic
were carried out in a blind way, without knowing the inflammation of enrolled children
results of any other examination. A positive result of
H. pylori infection was recorded when culture was positive
or when at least two out of three tests were positive. A Histological Analysis (Table 3)
negative result was recorded when all three tests were
concomitantly negative. Inflammation

Statistical Treatment Histological gastritis was detected in 53.95% antrum and in

This is a descriptive study; StatView software (Abacus,
CA, USA) was used for calculation of quantitative
parameters: mean, median, standard deviation (SD), and
range. Difference between the qualitative parameters was
calculated by using v2 test. Analysis of variance (ANOVA)
was also used for the crossing of qualitative and quantita-
tive parameters. P values \0.05 were considered
significant.
Results
Population and Endoscopic Findings
During the study period, only 619 children met the inclu-
sion criteria: 282 girls and 337 boys, with mean age
5.29 5.12 years (median 3.75 years, range 1 month to
17.75 years), with the age distribution displayed in Fig. 1.
The clinical purpose of endoscopy was RAP (51.3%)
children, GER (23.9%), vomiting (10.1%), failure to thrive
(6.7%), digestive hemorrhage (4.3%), and miscellaneous
(4.3%) (Table 1).
Endoscopic aspects were: normal (62.6%), congestion
(28.2%), nodular (6.9%) or erosive gastritis (1.6%). Only
three children had gastric ulcer and two bulbar ulcers
(Table 2).
59.12% corpus samples, mostly mild (grade 1). Children
with moderate (grade 2) histological gastritis were signif-
icantly older than those with grade 1 (P \ 0.02).
Activity
Histological active gastritis of grade 1 was detected in
18.57% antrum and corpus 20.03%. Children with an
active histological antro-corpus gastritis were significantly
older than those without active histological gastritis
(P \ 0.007).
Follicular Gastritis
Follicular gastritis was found in only 25 children (1.4%).
Gastric atrophy occurred in only one 13-year-old girl
who had suffered from GER since infancy. Following
persistence of recurrent epigastric pain and pyrosis, upper
digestive endoscopy highlighted presence of severe con-
gestive gastro-bulbitis. Antro-corpus biopsy specimens
evidenced H. pylori-negative moderate histological active
gastritis, associated with antral gastric atrophy. Following
3-month PPI treatment, she became completely asymp-
tomatic, and control endoscopic within 3 months showed
complete cure of all lesions.
Intestinal metaplasia occurred in a 15-year-old girl who
had suffered from GER since infancy. Also, following

123
Dig Dis Sci (2009) 54:19581965 1961

Table 1 Clinical features, histological inflammation, and ethnic origin according to H. pylori status
Clinical features H. pylori H. pylori H. pylori H. pylori H. pylori H. pylori
positive negative positive positive negative negative
66 (10.66) 553 (89.34) Antral Corpus Antral Corpus
n (%) n (%) inflammation inflammation inflammation inflammation
n (%) n (%) n (%) n (%)

RAP, n = 318 46 (14.46)
272 (85.63)* 38 (82.6)** 33 (71.7)** 152 (55.8) 151 (55.3)
GER, n = 148 6 (4.05) 142 (95.95)* 5 (83.3) 5 (83.3) 61 (42.9) 86 (60.5)
Vomiting, n = 63 6 (9.52) 57 (90.48)* 5 (83.3) 6 (100) 28 (49.1) 28 (49.1)
Failure to thrive, n = 42 3 (7.69) 39 (92.31)* 3 (100) 1 (33.3) 22 (56.4) 24 (61.5)
Digestive hemorrhages, n = 27 4 (14.81) 23 (85.79)* 3 (75) 2 (50) 10 (43.4) 15 (65.2)
Miscellaneous, n = 21 1 (4.77) 20 (95.23)** 0 0 9 (45) 17 (85%)
Ethnic origins
Caucasian, n = 516 42 (8.14) 474 (91.86)*
Non-Caucasian (Middle east 24 (23.31)* 79 (76.69)
or Africa), n = 103
* P \ 0.0001 according to the v2-test
** P 0.01 according to the v2-test


P 0.01 according to the v2-test, in children with RAP, H. pylori positive (46/66, 69.69%) versus H. pylori negative (272/553, 49.18%)

P 0.01 according to the v2-test, in children from the Middle east and Africa, H. pylori positive (24/66, 36.36%) versus H. pylori negative (79/
553, 14.28%)

persistence of recurrent epigastric pain and pyrosis, upper
digestive endoscopy showed severe congestive gastro-
bulbitis with erosive prepyloric lesions. Antro-corpus
biopsy specimens evidenced H. pylori-negative moderate
histological active gastritis, and low-grade intestinal
metaplasia located inside the cryptic antral epithelium.
Following 3-month PPI treatment, she became completely
asymptomatic and control endoscopic also showed disap-
pearance of all lesions.
H. pylori Status
Only 66 out of 619 children (10.66%) were H. pylori
positive (Table 4). All grade 2 and 3 H. pylori biopsy
specimens were concomitantly positive for bacterial cul-
ture and direct examination. In patients with grade 1
H. pylori, positive bacterial culture and direct examination
were seen in only one child in the antrum and two children
in the corpus (Table 4). H. pylori-positive children (antro-
corpus) were significantly older than H. pylori-negative
ones (7.4 4.75 years versus 5 5.1 years, P 0.0003).
Similarly, grade 3 H. pylori children were significantly
older than those with grades 2 and 1 (P \ 0.05).
Correlations of Histological, Clinical Manifestations,
Endoscopic Findings, and H. pylori Status
Caucasians (516, 83.3%) were significantly less infected
than non-Caucasians (Table 1).
Most clinical manifestations were significantly associ-
ated with H. pylori-negative status, but RAP was
significantly more frequent with H. pylori (Table 1). The
318 children with RAP exhibited significant antral and
corpus gastric inflammation [190 (59.75%) and 184
(57.86%), respectively; P \ 0.005 versus those without
histological antral and corpus inflammation] (Table 1).
However, children with RAP had a mild prevalence of
histological antral and corpus gastric activity: 94/318
children with RAP (29.55%) versus 224/318 (70.45%)
children without gastric activity (P \ 0.0001). The other
clinical manifestations were not associated with any spe-
cific histological features.
Antral histological gastritis and activity were signifi-
cantly higher in H. pylori-positive versus H. pylori-
negative children: 80.3 versus 50.8% and 71.2 versus
19.8% (P \ 0.001).
At endoscopy, H. pylori-negative children exhibited
significantly higher levels of normal esophageal, gastric,
and bulbo-duodenal mucosa than H. pylori-positive chil-
dren. Grade I and II oesophagitis, congestive, and erosive
gastritis were significantly more frequently detected in
H. pylori-negative children. Surprisingly, nodular gastritis
was detected in 60.46% of H. pylori-positive versus in
39.54% H. pylori-negative children (not significant, ns).
On the other hand, nodular gastritis was only detected in
17/553 H. pylori-negative children (3%) versus 26/66 of
H. pylori-positive children (39.3%) (P \ 0.0001). Gastric
ulcers were seen in only three H. pylori-negative children,

123
1962 Dig Dis Sci (2009) 54:19581965

Table 2 Endoscopical lesions according to H. pylori status Table 3 Gastric histological features according to H. pylori status
H. pylori H. pylori H. pylori H. pylori
positive negative positive negative
66 (10.66) 553 (89.34) 66 (10.66) 553 (89.34)
n (%) n (%) n (%) n (%)

Esophagus Antrum
Normal esophagus, n = 414 53 (12.8) 361 (87.2)* Inflammation
Inflammatory oesophagitis 10 (5.55) 170 (94.45)* Absence of inflammation, 13 (4.57) 272 (95.43)*
(grade I), n = 180 n = 285
Ulcerative oesophagitis 2 (8.7) 21 (91.3)** Grade 1, n = 278 18 (6.48) 260 (93.52)*
(grade II), n = 23 Grade 2, n = 47 28 (59.57) 19 (40.43)
Mycotic oesophagitis, n = 2 1 (50) 1 (50) Grade 3, n = 9 7 (77.77) 2 (22.23)
Endobrachy-esophagus, n = 1 0 1 (100) Activity
Oesophageal varices, n = 1 0 1 (100) Absence of activity, n = 462 19 (4.12) 443 (95.88)*
Stomach Grade 1, n = 115 24 (20.87) 91 (79.13)*
Normal stomach, n = 388 22 (5.68) 366 (94.32)* Grade 2, n = 42 23 (54.76) 19 (45.24)
Congestive gastritis, n = 175 17 (9.72) 158 (90.28)* Corpus
Nodular gastritis, n = 43 26 (60.46) 17 (39.54) Inflammation
Erosive gastritis, n = 10 1 (10) 9 (90)
Absence of inflammation, n = 253 20 (7.91) 233 (92.09)*
Gastric ulcer, n = 3 0 3 (100) Grade 1, n = 309 16 (5.18) 293 (94.82)*
Bulb Grade 2, n = 55 28 (50.9) 27 (49.1)
Normal bulb, n = 557 54 (9.69) 503 (90.31)* Grade 3, n = 2 2 (100) 0
Congestive bulbitis, n = 26 3 (11.53) 23 (88.47)* Activity
Nodular bulbitis, n = 22 6 (27.27) 16 (72.73) Absence of activity, n = 454 26 (5.73) 428 (94.27)*
Atrophic bulbitis, n = 12 2 (16.66) 10 (83.34) Grade 1, n = 124 21 (17) 103 (83)*
Bulbar ulcer, n = 2 1 (50) 1 (50) Grade 2, n = 41 19 (46.34) 22 (53.66)
Duodenum Antrum and corpus
Normal duodenum, n = 614 65 (10.58) 549 (89.42)* Follicular gastritis
Congestive duodenitis, n = 3 1 (33.33) 2 (66.67) Absence of follicular 43 (7.22) 553 (92.78)*
Nodular duodenitis, n = 2 0 2 (100) gastritis, n = 596
* P \ 0.0001 according to the v2-test Presence of follicular 23 (92)* 2 (8)
gastritis, n = 25
** P 0.002 according to the v2-test


P 0.04 according to the v2-test Note: One girl presented with H. pylori-negative mild gastric atrophy
 2 and another one with H. pylori-negative mild intestinal metaplasia
P 0.05 according to the v -test
* P \ 0.001 according to the v2-test

and bulbar ulcers in one H. pylori-positive child and one Discussion

H. pylori-negative child (Table 2). Antro-corpus histolog-
ical inflammation was seen in 68.57% of children with
congestive gastritis and in 86.4% of those with nodular
gastritis. Comparatively, antro-corpus histological activity
was detected in only 31.15% and 43.49%, respectively.
Finally, none of the other endoscopic aspects exhibited
any significant difference of histological gastritis or
activity.
Low grades (mild) of gastric inflammation and activity
were both associated with H. pylori-negative status. Fol-
licular gastritis in antrum and corpus was significantly
associated with H. pylori positivity (Table 3). Children
with antro-corpus histological follicular gastritis were sig-
nificantly older (9.75 4.5 years versus 5.1 4.9 years,
P \ 0.0001).
This study suggests that, in children referred with symp-
toms of clinical gastritis, endoscopy reveals that primary
mild-grade histological gastritis is frequent, amounting to
53% in the antrum and 59% in the corpus, that H. pylori
infection is infrequent, detected in only 10.6% of cases,
and that histological gastritis appears largely unrelated to
H. pylori infection. In addition, both gastric atrophy and
intestinal metaplasia are rare.
The population evaluated in this study was characterized
by symptoms of clinical gastritis represented mainly by
RAP, GER, and vomiting, usually leading to upper GI
endoscopy according to the recommendation of the North
American Society of Pediatric Gastroenterology, Hepatol-
ogy, and Nutrition (NASPGHAN) [5, 14, 15], so that it is

123
Dig Dis Sci (2009) 54:19581965 1963

Table 4 The results of histology, bacterial culture, and direct tool. Interestingly, expressing the endoscopic features in
examination according to H. pylori status relation to the 66 H. pylori-positive or 553 H. pylori-neg-
H. pylori H. pylori ative children shows that nodular gastritis is not as frequent
positive negative in our population: 39.3% versus 3% of cases, respectively.
66 (10.66) 553 (89.34) Besides, also that normal stomach feature is present in 33%
n (%) n (%)
of H. pylori-positive and in 66% of H. pylori-negative
Histology children. Furthermore, H. pylori-negative children exhib-
H. pylori (antrum) ited significantly higher levels of normal esophageal,
Absence of H. pylori, n = 511 4 (0.78) 507 (99.22)* gastric, and bulbo-duodenal mucosa than H. pylori-positive
Grade 1, n = 47 1 (2.12) 46 (97.88)*
children. These endoscopic findings could probably be due
Grade 2, n = 30 30 (100) 0
to a certain patient selection bias.
Peptic ulcer in childhood is exceptional and H. pylori
Grade 3, n = 31 31 (100) 0
gastritis without peptic ulcer is frequent in children. In a
H. pylori (corpus)
previous study, one gastric and two duodenal ulcers were
Absence of H. pylori, n = 507 14 (2.76) 493 (97.24)*
found in 47 children with H. pylori infection [3]. Further-
Grade 1, n = 56 2 (3.57) 54 (96.43)*
more, in a recent prospective European multicenter pilot
Grade 2, n = 30 30 (100) 0
study on the incidence in children of gastric and duodenal
Grade 3, n = 26 26 (100) 0
ulcer disease or erosions, ulcers occurred in 10.6% of
Bacterial culture
cases, with H. pylori infection in only 26.7% of these [20].
Negative, n = 553 0 553 (100)
In agreement with these previous data, only 1 out of 66
Positive, n = 66 66 (100) 0
H. pylori-positive children exhibited bulbar ulcer.
Direct examination
Antral and corpus histological gastritis were found in
Negative, n = 553 0 553 (100)
53.95% and 59.12% of cases and most were H. pylori
Positive, n = 66 66 (100) 0 negative. In previous pediatric studies, prevalence of his-
2
* P \ 0.0001 according to v -test tological chronic gastric inflammation varied from 75% to
100% [25]. The lowest prevalence of chronic gastritis was
reported by American pediatric series: 32% for Elitsur
likely to encompass any type of disease affecting the upper et al. [21] and 26.31% for Mahony et al. [22]. Chronic
digestive tract, especially gastritis in its different aspects. gastritis occurs in boys as well as in girls [4, 10]. Fur-
Endoscopy showed normality or congestion in more than thermore, the frequency of gastritis increases in correlation
90% of cases. Nodular gastritis, the aspect most clearly with age [5, 14, 21], in agreement with our results. The
related to H. pylori infection, found in up to 53% of cases classical aspect of chronic active gastritis commonly
[3, 16, 17], was seen in only 60.46%. Nodular gastritis is not observed in adults, characterized by prevalence of poly-
absolutely specific of H. pylori infection and nodular gas- nuclear neutrophils in the inflammatory infiltrate, is rarer in
tritis cases without H. pylori have already been reported [3, children [7, 23, 24], also in agreement with our study.
17, 18]. Helicobacter heilmannii infection may be involved; Chronic follicular gastritis was suggested to be specific
it is usually large and very distinctive when seen in histo- to H. pylori infection [1, 2, 10]. Chronic follicular gastritis
pathologic sections, and thus should have been reported by corresponds to a diffuse and polymorphic inflammatory
the pathologists reviewing these cases [19]. However, infiltrate comprising a very high number of lymphoid fol-
probably in a considerable number of cases, nonspecific licles in a gastric mucosa [1]. Those data are in agreement
antibiotic treatments for other nonspecific infections used with our study since follicular gastritis was significantly
previous to endoscopy could decrease the concentration of more frequent in the antrum and corpus of H. pylori-
the bacteria on mucous membrane surfaces, hence leading positive children.
to false-negative results. In our study, surprisingly, nodular In adults, gastric glandular atrophy and/or intestinal
gastritis was not significantly associated with H. pylori metaplasia are frequently associated with H. pylori infec-
even though all enrolled children had been free of any tion and both are considered preneoplastic lesions [12, 13].
antibiotics, anti-H2 or PPI use for at least the preceding This relationship is still controversial in the pediatric
4 weeks, and the search for H. heilmannii infection was also population and the incidence of gastric atrophy and/or
negative. Furthermore, this nodular endoscopic aspect intestinal metaplasia is not consistent among published
could have been related to former H. pylori infection, since studies [2, 13, 2430]. The absence of consistence rela-
this endoscopic feature persists a long time after proper tionship between H. pylori infection and gastric glandular
eradication of the bug and the only way to decide would atrophy and/or intestinal metaplasia in the reported pedi-
have been by using an immunoblot H. pylori diagnostic atric studies is might be due to the fact that H. pylori

123
1964
serology was not performed to define if there was past
infection [2, 13, 2430]. In fact, diagnosis of gastric
glandular atrophy as well as intestinal metaplasia in chil-
dren is very difficult because it requires an adequate
number of properly oriented gastric biopsy specimens;
furthermore, those lesions have a patchy gastric distribu-
tion. Finally, interobserver agreement is poor, especially in
the assessment of atrophy. At present gastric glandular
atrophy and intestinal metaplasia criteria in children are not
established and are urgently needed [27, 30].
In Tunisia, Maherzi et al. [2] reported a frequency of
gastric atrophy in 25.38% of children, always associated
with H. pylori. In the USA, Guarner et al. [13] reported
gastric atrophy and/or intestinal metaplasia in 12 of 19
(63%) children infected with H. pylori, gastric atrophy
alone being observed in eight, intestinal metaplasia alone
in two, and both in two. In Turkey, Usta et al. [27] reported
gastric atrophy and/or intestinal metaplasia in only 5/175
H. pylori-positive children (2.8%), atrophy alone in three,
intestinal metaplasia in one, and both lesions in one. Those
results contrast with those of another Turkish study
showing glandular atrophy and/or intestinal metaplasia in
14 of 18 (77.7%) H. pylori-positive children [28]. Other
researchers did not observe any atrophic gastritis in chil-
dren [2426]. Finally in France, Blain-Stregloff et al. [29]
reported a frequency of gastric atrophy in 5% of cases and
Trinh et al. [31] in only one case, in close agreement with
our data.
Intestinal metaplasia is rare in children. It is the conse-
quence of prolonged H. pylori infection and its prevalence
increases with time [32]. Shabib et al. [33], comparing two
groups of children with H. pylori-positive gastritis versus
H. pylori-negative gastritis, reported a frequency of intes-
tinal metaplasia, respectively, in 42% and 6% of cases. On
the other hand, Blain-Stregloff et al. [29] reported only one
case (0.3%) of intestinal metaplasia, whereas Ilboudo et al.
[25] and Chong et al. [26] did not observe any cases.
This study shows only 10.66% of children with
H. pylori infection, in good agreement with the study
carried out by Wizla-Derambure et al. [34] with the pur-
pose of identifying familial and community environmental
risk factors associated with H. pylori infection in a pedi-
atric population requiring diagnostic upper endoscopy
during a 2-year period. The authors included 436 patients
(242 boys), aged 2 days to 17.9 years (median 2.7 years),
and H. pylori was present in 7.3% [34].
The low incidence of H. pylori infection in our popu-
lation contrasts with a high level of H. pylori-negative
histological gastritis, up to almost two out of three cases.
For Hessey et al. [35], absence of H. pylori during gastritis
could be related to the local inflammatory answer itself,
which in certain cases would lead to elimination of the
germ in contact with the mucous membrane, hence
123
Dig Dis Sci (2009) 54:19581965
transient persisting histological gastritis. According to
another hypothesis, histologic gastritis could be related to
infection by other H. pylori species not detected by con-
ventional methods, i.e., histology and culture. Further
molecular analysis of the gastric biopsy specimens could
be helpful [36].
The exact relationship between H. pylori infection and
macroscopic and histologic gastritis still remains uneluci-
dated. In a previous study [37], we found an extremely high
prevalence of macroscopic and histologic gastritis (71.7%)
in noninfected children [37]. This highly selected popula-
tion of children with epigastric pain was actually close to
that depicted by Snyder et al. [10], where PAG was found
in about 20% of the four different age groups of 408
children, with only 4/39 children \10 years old with PAG
having evidence of H. pylori infection. Trinh et al. [31]
compared prospectively two groups of children, one with a
histological gastritis and the other one with a normal his-
tology, with no significant difference between the two
groups with regard to existence of digestive symptoms. In
our study, RAP was significantly but only slightly higher in
H. pylori-positive children. Furthermore, children with
RAP were significantly associated with H. pylori-negative
nonactive antro-corpus histological gastritis.
Conclusion
H. pylori infection was rare in this systematic analysis of
endoscopic features in children with clinical gastritis, but
associated with (1) a high level of mild-grade histological
antro-corpus primary gastritis and activity and (2) a sig-
nificantly higher rate of follicular in gastric antrum and
corpus. Intensity and activity of histological gastritis both
increased with age. Nodular and congestive gastritis were
associated with nonactive histological gastritis. Gastric
atrophy and intestinal metaplasia seem rare in children.
RAP were more frequent in case of H. pylori infection.
References
1. Dixon MF, Genta RM, Yardley JH, Correa P. Classification and
grading of gastritis. The update Sydney System. International
workshop on the histopathology of gastritis. Houston 1994. Am J
Surg Pathol. 1996;20:11611181. doi:10.1097/00000478-1996
10000-00001.
2. Maherzi A, Fendri C, Ben Jilani S, Bousnina S. Infection
symptomatique a` Helicobacter pylori: etude prospective des
aspects epidemiologiques, diagnostiques et therapeutiques chez
lenfant en Tunisie. Arch Pediatr. 1996;3:329334. doi:10.1016/
0929-693X(96)84686-0.
3. Benhamou PH, Kalach N, Raymond J, Abdallah C, Dupont C.
Infections gastriques a` Helicobacter pylori chez lenfant. Presse
Med. 1994;23:17031707.
Dig Dis Sci (2009) 54:19581965
4. Gottrand F, Cullu F, Turck D, et al. Normal gastric histology in
Helicobacter pylori infected children. J Pediatr Gastroenterol
Nutr. 1997;25:7478. doi:10.1097/00005176-199707000-00012.
5. Guiraldes E, Pena A, Duarte I, et al. Nature and extent of gastric
lesions in symptomatic Chilean children with Helicobacter
pylori-associated gastritis. Acta Paediatr. 2002;91:3944. doi:
10.1080/080352502753457923.
6. Drumm B, Sherman P, Chiasson D, Karmali M, Cutz E. Treat-
ment of Campylobacter pylori-associated antral gastritis in
children with bismuth subsalicylate and ampicillin. J Pediatr.
1988;113:908912. doi:10.1016/S0022-3476(88)80030-1.
7. Drumm B, Sherman P, Cutz E, Karmali M. Association of
Campylobacter pylori on the gastric mucosa with antral gastritis
in children. N Engl J Med. 1987;316:15571561.
8. Dooley CP, Cohen H, Fitzgibbons PL, et al. Prevalence of Hel-
icobacter pylori infection and histologic gastritis in
asymptomatic persons. N Engl J Med. 1989;321:15621566.
9. Prieto G, Polanco I, Larrauri J, Rota L, Lama R, Carraszo S.
Helicobacter pylori infection in children: clinical, endoscopic and
histologic correlations. J Pediatr Gastroenterol Nutr. 1992;14:
420425. doi:10.1097/00005176-199205000-00008.
10. Snyder JD, Hardy SC, Thorne GM, Hirsch BZ, Antonioli DA.
Primary antral gastritis in young American children. Low prev-
alence of Helicobacter pylori infection. Dig Dis Sci. 1994;39:
18591863. doi:10.1007/BF02088115.
11. Flejou JF, Diomade I, Molas G, et al. Gastrite chronique associee
chez lhomme a` la presence de germes spirales non Helicobacter
pylori. Gastroenterol Clin Biol. 1990;14:806810.
12. Guarner J. The spectrum of gastric disease associated with Heli-
cobacter pylori and other infectious gastritides. Curr Gastroenterol
Rep. 2004;6:441446. doi:10.1007/s11894-004-0065-7.
13. Guarner J, Bartlett J, Whistler T, et al. Can pre-neoplastic lesions
be detected in gastric biopsies of children with Helicobacter
pylori infection? J Pediatr Gastroenterol Nutr. 2003;37:309314.
doi:10.1097/00005176-200309000-00019.
14. Squires RH, Coletti RB. Indications for paediatric gastrointestinal
endoscopy: a medical position statement of the North American
Society for Paediatric Gastroenterology and Nutrition. J Pediatr
Gastroenterol Nutr. 1996;23:107111. doi:10.1097/00005176-
199608000-00002.
15. Bahu MGS, Silveira TR, Maguilnick I, Ulbrich-Kulezynski J.
Endoscopic nodular gastritis: an endoscopic indicator of high-
grade bacterial colonization and severe gastritis in children with
Helicobacter pylori. J Pediatr Gastroenterol Nutr. 2003;36:217
222. doi:10.1097/00005176-200302000-00011.
16. Cadranel S, Goossens H, De Boeck M, Malengream H, Rodesch
P, Butzler JP. Campylobacter pyloridis in children. Lancet.
1986;1:735736. doi:10.1016/S0140-6736(86)91120-7.
17. Oderda G, Lerro P, Poli E, Tavassali K, Ansaldi N. Childhood
nodular gastritis and Campylobacter pylori. Endoscopy.
1988;20:86.
18. Kalach N, Benhamou PH, Raymond J, et al. Infections gastriques
a` Helicobacter pylori chez lenfant. Presse Med. 1997;26:1688
1694.
19. Morgner A, Bayerdroffer E, Meining A, Stolte M, Kroher G.
Helicobacter heilmannii and gastric cancer. Lancet. 1995;346:
511512. doi:10.1016/S0140-6736(95)91364-5.
20. Kalach N, Bontems P, Mourad-Baars P, et al. Prospective
European multi-centre pilot study on gastric and duodenal ulcer
disease or erosions in children. Helicobacter. 2007;12:394.
1965
21. Elitsur Y, Raghuverra A, Sadat T, Vaid P. Is gastric nodularity a
sign for gastric inflammation associated with Helicobacter pylori
infection in children? J Clin Gastroenterol. 2000;30(3):286288.
doi:10.1097/00004836-200004000-00016.
22. Mahony MJ, Wyatt JI, Littlewood JM. Campylobacter pylori
gastritis. Arch Dis Child. 1988;63:654655.
23. De Giacomo C, Fiocca R, Villani L, et al. Helicobacter pylori
infection and chronic gastritis: clinical, serological and histologic
correlations in children treated with amoxicillin and collodal
bismuth subcitrate. J Pediatr Gastroenterol Nutr. 1990;11:310
316. doi:10.1097/00005176-199010000-00005.
24. Cohen MC, Cueto Rua E, Balcaree N, et al. Assessment of the
Sydney system in H. pylori-associated gastritis in children. Acta
Gastroenterol Latinoam. 2000;30:3540.
25. Ilboudo D, Bougouma A, Sombie R, et al. Infection a` Helico-
bacter pylori chez lenfant en zone tropicale. Gastroenterol Clin
Biol. 1998;22:855857.
26. Chong SKF, Lou Q, Asnicar MA, et al. Helicobacter pylori in
recurrent abdominal pain in childhood: comparison of diagnostic
tests and therapy. Pediatrics. 1995;96:211215.
27. Usta Y, Saltik-Temizel IN, Ozen H. Gastric atrophy and intestinal
metaplasia in Helicobacter pylori infection. J Pediatr Gastroen-
terol Nutr. 2004;38:548. doi:10.1097/00005176-200405000-
00018.
28. Ozturk Y, Buyukgediz B, Arslan N, Ozer E. Antral glandular
atrophy and intestinal metaplasia in children with H. pylori
infection. J Pediatr Gastroenterol Nutr. 2003;37:9697. doi:
10.1097/00005176-200307000-00020.
29. Blain-Stregloff AS, Mention K, De Martin LassalleE, et al.
Atrophie gastrique et metaplasie intestinale chez lenfant. Arch
Pediatr. 2003;10:807808.
30. Dimitrov G, Gottrand F. Does gasdtric atrophy exist in children?
World J Gastroenterol. 2006;12:62746279.
31. Trinh N, Tounian P, Josset P, et al. Etude anatomo-clinique
prospective des gastrites de lenfant. Les gastrites chroniques
chez lenfant. Arch Pediatr. 2001;8:106.
32. Peuchmaur M, Faure C, Mougenot JF, et al. Gastrites chroniques
associees a` Helicobacter pylori chez lenfant. Ann Pathol.
1994;14:315318.
33. Shabib SM, Cutz E, Drumm B, et al. Association of gastric
metaplasia and duodenitis with Helicobacter pylori infection in
children. Am J Clin Pathol. 1994;102:188191.
34. Wizla-Derambure N, Michaud L, Ategbo S, et al. Familial and
community environmental risk factors for Helicobacter pylori
infection in children and adolescent. J Pediatr Gastroenterol
Nutr. 2001;33:5863. doi:10.1097/00005176-200107000-00010.
35. Hessey SJ, Spencer J, Wyatt JI, et al. Bacterial adhesion and
disease activity in Helicobacter pylori associated chronic gastri-
tis. Gut. 1990;31:134138. doi:10.1136/gut.31.2.134.
36. Nilsson H, Olsson G, Lundkvist A, Ljungh A, Wadstrom T.
Enterohepatic Helicobacter species in Swedish wild rodents.
Helicobacter. 2007;12:383.
37. Kalach N, Mention K, Guimber D, Michaud L, Spyckerelle C,
Gottrand F. Helicobacter pylori infection is not associated with
specific symptoms in non-ulcer dyspeptic children. Pediatrics.
2005;115:1721.
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.