You are on page 1of 6

Journal of Pediatric Gastroenterology and Nutrition

36:217–222 © February 2003 Lippincott Williams & Wilkins, Inc., Philadelphia

Endoscopic Nodular Gastritis: An Endoscopic Indicator of

High-Grade Bacterial Colonization and Severe Gastritis in
Children With Helicobacter pylori
*Maria da Graça Soares Bahú, †Themis Reverbel da Silveira, ‡Ismael Maguilnick, and
§Jane Ulbrich-Kulczynski
*Pediatric Gastroenterology Unit, Hospital da Criança Conceição, Federal Department of Health, †Pediatric Gastroenterology
Unit and ‡Digestive Endoscopy Unit, Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, and
§Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, and Pathology Department, Hospital da
Criança Conceição, Federal Department of Health, Porto Alegre, Brazil

ABSTRACT positive predictive value (91.7%) for the diagnosis of H. pylori

Objective: To investigate the significance of endoscopic nodu- infection and was observed in 22 of 50 (44%) H. pylori-positive
lar gastritis associated with Helicobacter pylori infection. patients and in 2 of 135 (1.5%) H. pylori-negative patients. A
Methods: This prospective study included 185 children (50.8% significant association was observed between older age and the
boys) aged 1 to 12 years (mean, 6.9 ± 3.0 years) who underwent prevalence of this finding (P < 0.001). There was a significant
upper intestinal endoscopy during evaluation of chronic ab- increase in endoscopic nodular gastritis with increased H. py-
dominal pain. The authors assessed the endoscopic appearance lori density and a positive correlation (Pearson coefficient ⳱
of the stomach, noting those patients with endoscopic nodular 0.97) with increased gastritis score on histologic examination.
gastritis. Urease activity of gastric mucosal biopsies was mea- Increase in gastritis score was dependent on increased H. pylori
sured. With histologic examination, the presence and density of density in patients with gastric nodularity; this finding was
H. pylori organisms, the presence of follicular gastritis, the independent of age.
nature of inflammation, and the gastritis activity grade and Conclusions: Endoscopic findings of antral nodularity in chil-
overall gastritis score were assessed. dren suggest the presence of H. pylori infection and follicular
Results: H. pylori infection was identified in 50 children gastritis and may identify cases of severe gastritis and marked
(27%). Endoscopic nodular gastritis was significantly associ- bacterial colonization. JPGN 36:217–222, 2003. Key Words:
ated with active chronic gastritis and follicular gastritis. Nodu- Helicobacter pylori—Gastritis—Endoscopy—Abdominal
larity in the stomach showed a high specificity (98.5%) and pain. © 2003 Lippincott Williams & Wilkins, Inc.

Helicobacter pylori infection is a common and uni- tional level of the head of the family are also important
versally distributed bacterial infection. It is predomi- risk factors for infection (5,6).
nantly acquired in childhood (1), and more than three H. pylori infection has fulfilled Koch postulates (7)
fourths of the population in developing countries is in- and met Hill criteria for being a cause of active chronic
fected during childhood. In developed countries, infec- gastritis (8). It is the most frequent cause of chronic
tion in children is less frequent (2). The major risk factor gastritis and peptic ulcer, and is directly associated with
for H. pylori infection is crowded living conditions (1,3– gastric carcinoma and mucosa-associated lymphoid tis-
5), but epidemiologic studies have shown that poor sani- sue (MALT) lymphoma (1).
tation, lack of adequate water supply, and a low educa- The inflammation that H. pylori causes in the gastric
mucosa is not always observed macroscopically with en-
Received April 11, 2002; accepted September 30, 2002.
Supported in part by Coordenação de Aperfeiçoamento de Pessoal de doscopy, but it is identified on the histologic examination
Ensino Superior (CAPES), Brazil. of gastric biopsies (9–11). Endoscopic findings of nodu-
Technical support was provided by the Graduate Program in Gas- larity can be seen in the stomach of children much more
troenterology, Universidade Federal do Rio Grande do Sul, and the frequently than in adults. This nodularity, more often
Graduate Research Group, Hospital de Clínicas de Porto Alegre, Brazil.
Address correspondence and reprint requests to Maria da Graça observed in the gastric antrum, has been called antral
Soares Bahú, Ildefonso Simões Lopes, 201/02, 91330–180 Porto nodular gastritis (also, nodular antritis or gastric lympho-
Alegre, RS, Brazil (e-mail: nodular hyperplasia) (10,12–17). Some authors (12,13,


18) believe that this macroscopic change may predict H.

pylori infection and histologic gastritis. Other authors,
however, still question the specificity of endoscopic
nodular gastritis for H. pylori infection (19).
We performed a cross-sectional study of children with
chronic abdominal pain to investigate the association of
H. pylori infection with endoscopic nodular gastritis and
histologic features of gastritis.


Patients and Samples

One hundred eighty-five children (50.8% boys) aged 1 to 12

years (mean, 6.9 ± 3.0 years) were included in this prospective
study performed from 1997 to 1999 in two hospitals in Porto
Alegre, Brazil. All the children came from low-income fami-
lies, and the level of education of their parents was low. They
all presented with chronic abdominal pain and were referred for
upper intestinal endoscopy. Only the children whose parents
gave informed consent participated in the study. Exclusion cri-
teria were contraindication for endoscopy; use of antibiotics,
bismuth, H2 blockers, or proton pump inhibitors within 30 days FIG. 1. Nodularity seen in the gastric antrum during endoscopic
of the procedure; and use of antiinflammatory drugs or acetyl- examination.
salicylic acid (ASA) within 3 days of the procedure. Patients
taking immunosuppressive agents or chemotherapy drugs were
Histologic Examination
excluded, as were patients with immunodeficiency or suspected
celiac disease. Four age groups were established: 1 to 3 years
One of the authors (J.U.K.), blinded to the patients’ endo-
(35 patients), 4 to 6 years (58 patients), 7 to 9 years (48 pa-
scopic or clinical findings, examined the biopsies (stained with
tients), and 10 to 12 years (44 patients).
hematoxylin and eosin, and Giemsa) for H. pylori and gastritis.
Endoscopy was performed by one of the authors (M.G.S.B.).
H. pylori infection was characterized by bacteria found with
An Olympus GIF 130 videoendoscope (outer diameter ⳱ 9.5
mm; biopsy channel diameter ⳱ 2.8 mm) was used combined
with EVIS CV 100 video processor and FB-25K biopsy for-
ceps. Five gastric biopsy specimens were collected from each
patient for histologic analysis: two from the gastric antrum (one
from the distal lesser curvature and one from the distal greater
curvature), two from the corpus (one from the lesser curvature
and one from the greater curvature), and one from the area next
to the incisura angularis. Histologic examination followed the
Sydney System criteria as updated in Houston in 1994 (20).
Specimens were placed directly in 10% formalin without a
paper filter. A sixth specimen from the gastric antrum was
collected for the urease rapid test (21). This section was em-
bedded in GASTROtest gel (Santa Cruz do Sul, Brazil), which
consists of a small portion of modified Christensen urea agar
and a pH indicator.

Endoscopic Diagnosis

The diagnosis of endoscopic nodular gastritis was made

when the mucosa had an irregular appearance resembling a
cobblestone pavement. Micronodules measured 1 to 4 mm in
diameter, had a smooth surface, and were the same color as the
surrounding mucosa, as described by Hassal and Dimmick (22)
in 1991 (Fig. 1). These nodules, evident with introduction of
the endoscope, were better identified after biopsy collection
because the blood from the biopsy site surrounded and high- FIG. 2. Evident nodularity seen after collection of endoscopic
lighted them (Fig. 2). biopsies.

J Pediatr Gastroenterol Nutr, Vol. 36, No. 2, February 2003


histologic examination, and density of organisms was rated increased H. pylori density in the gastric mucosa (Spear-
low, moderate, or high. The diagnosis of gastritis was based on man correlation coefficient, 0.58).
the histopathologic findings of inflammation, activity, metapla- The diagnostic performance of the urease test for H.
sia, and atrophy. Each of these findings was scored as normal pylori infection in 176 patients was as follows: sensitiv-
(0), mild (1), moderate (2), or marked (3) following the updated ity, 60.4%; specificity, 94.5%; positive predictive value,
Sydney System and using visual analog scales (20). The final
gastritis score was the sum of the four histopathologic scores.
80.5%; and negative predictive value, 83%.
Finally, the diagnosis of follicular gastritis was based on the Active chronic gastritis was significantly more preva-
presence of lymphoid follicles with a germinative center and a lent in patients with H. pylori infection (24 of 50; 48.0%)
mononuclear infiltrate in the lamina propria of the gastric mu- than in patients not infected (6 of 135; 4.4%); ␹2 ⳱ 47.8
cosa. (P < 0.001); PR ⳱ 10.8 (95% CI, 4.7–24.9). Prevalence
increased significantly with the increase of bacterial den-
sity; ␹2 ⳱ 84.8 (P < 0.001). Spearman correlation coef-
Ethical Considerations and Variables Studied
ficient calculated to assess the association of H. pylori
The Scientific Committee and the Health Research Ethics density and severity of gastritis with increase in age was
Committees of the Department of Research and Graduate Stud- 0.17 and 0.07, respectively.
ies at Hospital de Clínicas de Porto Alegre and Hospital Con- Endoscopic nodular gastritis was seen in 24 of 185
ceição (Porto Alegre, Brazil) approved this research project. patients (13%; 95% CI, 8.5–18.7). Of these, 58.3% were
The variables studied were demographics, endoscopic nodu- boys. An association between older age groups and the
lar gastritis, urease test findings, H. pylori infection, follicular prevalence of this finding was observed (P < 0.001;
gastritis, inflammation (mononuclear cell infiltration) and gas- Table 1). Endoscopic nodular gastritis was found in 22 of
tritis activity (polymorphonuclear cell infiltration) grades, H.
the 50 (44%) patients with H. pylori infection and in only
pylori density found with histologic examination, and gastritis
score. 2 of the 135 (1.5%) who were H. pylori negative; ␹2 ⳱
54.72 (P < 0.001); PR ⳱ 29.7 (95% CI, 7.3–121.7).
Endoscopic nodular gastritis was significantly associ-
Statistical Analysis ated with active chronic gastritis identified during histo-
logic examination (15 of 30 [50%] vs. 9 of 155 [5.8%];
Results were expressed as number and percentage. Age was
Fisher exact test: P < 0.001; PR ⳱ 8.6; 95% CI, 4.2–
described as median and range. Percentage comparisons were
assessed using the ␹2 test or Fisher exact test when appropriate. 17.8) and with follicular gastritis (12 of 24 [50.0%] vs.
We also calculated prevalence ratio (PR) and 95% confidence 12 of 161 [7.5%]; ␹2 ⳱ 29.8 [P < 0.001]; PR ⳱ 6.7;
interval (CI). Correlation coefficients (Pearson and Spearman) 95% CI, 3.4–13.2).
were used for specific comparisons involving age, H. pylori The diagnostic performance of endoscopic nodular
density, and gastritis score. Combined effects were assessed gastritis for H. pylori infection was sensitivity, 44%;
using the logistic regression model to obtain adjusted odds ratio specificity, 98.5%; positive predictive value, 91.7%; and
for the association of age, H. pylori density, and gastritis score negative predictive value, 82.5%. Table 1 shows the sig-
with gastric nodularity. Significance was set at 0.05. Data were nificant increase in endoscopic nodular gastritis inci-
processed and analyzed with SPSS version 8.0 (SPSS Inc., dence with increased H. pylori density.
Chicago, IL, U.S.A.).
Endoscopic nodular gastritis was significantly associ-
ated with higher grades of gastritis severity observed
TABLE 1. Prevalence of endoscopic nodular gastritis
H. pylori infection was identified in 50 (27%) of the according to age and H. pylori density
185 children (95% CI, 20.8–34.0). There were 26 boys,
and age ranged from 1.7 to 12 years (median, 8.5 years). Endoscopic nodular gastritis
Bacterial density found with histologic examination was Variable Number Frequency Percent PR 95% CI
low in 23 (46%) of the 50 infected patients, moderate in
Age group*
12 patients (24%), and high in 15 (30%). There was a 1–3 y 35 0 0 — —
significant increase in H. pylori infection in association 4–6 y 58 4 6.9 5.5 0.3–99.0
with age: ␹2 ⳱ 5.35 (P ⳱ 0.021). All 50 children in- 7–9 y 48 8 16.7 12.5 0.7–209.5
fected with H. pylori had chronic gastritis (mononuclear 10–12 y 44 12 27.3 20.0 1.2–785.2
cell infiltration) found on histologic examination. Of Total 185
H. pylori density†
these, 24 had active chronic gastritis (polymorphonuclear Absent 135 2 1.5 — —
cell infiltration). Low 23 6 26.1 17.6 3.8–82.0
With regard to gastritis severity, 12 of 50 H. pylori- Moderate 12 6 50.0 33.8 7.6–149.4
positive children had a score of 1 (24%), 9 had a score of High 15 10 66.7 45.0 10.9–186.4
2 (18%), 11 had a score of 3 (22%), 9 had a score of 4 * PR ⳱ adjusted prevalence ratio based on Agresti’s correction: 0.5;
(18%), 7 had a score of 5 (14%), and 2 had a score of 6 ␹2 for linear trend ⳱ 15.4 (P < 0.001). †PR ⳱ prevalence ratio; ␹2 for
(4%). An increase in gastritis score was observed with linear trend ⳱ 71.5 (P < 0.001).

J Pediatr Gastroenterol Nutr, Vol. 36, No. 2, February 2003


during histologic examination and determined according TABLE 2. Logistic regression model representing odds
to gastritis score. Pearson correlation coefficient was ratio for the association of age, H. pylori density, and
0.97 (P < 0.001). Figure 3 shows, according to gastritis gastritis score with gastric nodularity
score, the prevalence of nodularity in the gastric antrum Odds ratio 95% CI P
during endoscopy.
The odds ratio obtained with the logistic regression Age 1.03 1.01–1.05 0.004
H. pylori density 3.86 1.86–8.00 <0.001
model for the association of age, H. pylori density, and Gastritis score 1.31 0.74–2.32 0.353
gastritis score with gastric nodularity can be seen in
Table 2. Mutual adjustments of variables showed that
increased H. pylori density and age were independently density in the gastric sample, which is usually lower in
associated with presence of gastric nodularity. Increase children than in adolescents and adults (10,26,27). This
in gastritis score was dependent of increased H. pylori difference may be the reason for the low sensitivity of
density and significantly associated with gastric nodular- our method.
ity. In 1982, Warren and Marshal described H. pylori in
the gastric mucosa and reported the presence of endo-
scopic nodular gastritis in positive cases (28). Later,
DISCUSSION Czinn et al. (29) and Cadranel et al. (30) reported the
same endoscopic finding in children. Nodularity was as-
The children included in this study had some of the sociated with lymphoid follicles seen with histologic ex-
common risk factors for H. pylori infection, such as low amination in some of their patients. More recent studies
family income and parental education level. Of these have also reported a significant association between en-
patients, 27% had H. pylori infection, and infection in- doscopic nodular gastritis and H. pylori infection (12–
creased with age. This result is the same found in reports 15,18,31,32).
by several authors (3,4,15,23). We also found a significant association between en-
The urease test used in this study (GASTROtest, Santa doscopic nodular gastritis and H. pylori infection (as
Cruz do Sul, Brazil) is a commercial test with high speci- seen in 44% of the patients with H. pylori infection and
ficity (94.5%) but low sensitivity (60.4%). This feature is in only 1.5% of the patients without infection). Our re-
somewhat different than the commercial CLOtest (Delta sults are similar to those reported by Luzza et al. (45.3%
West Bertley, Australia), whose sensitivity is reported to for H. pylori-positive patients and 1.5% for H. pylori-
be 70% to 85% (24–26). In accordance with the medical negative patients) and by De Giacomo et al. (43% for H.
position statement of the North American Society for pylori-positive patients and 4% for H. pylori-negative
Pediatric Gastroenterology and Nutrition, we believe that patients) (14,32).
the accuracy of the test depends mainly on the H. pylori Specificity (98.5%) and positive predictive value
(91.7%) of endoscopic nodular gastritis for H. pylori
infection were high, which confirms the findings of Pri-
eto Bozano et al. (18). The association between endo-
scopic nodular gastritis and follicular gastritis was ob-
served in our study and is supported by findings de-
scribed by other authors (12,14,18). Therefore, we can
say that positive findings for endoscopic nodular gastritis
point to a high probability of H. pylori infection because
the positive predictive value is high (27.0%) at low
prevalence rates of the disease. However, the absence of
nodularity in the gastric mucosa on histologic examina-
tion does not exclude the possibility of H. pylori infec-
Low bacterial density in the gastric mucosa and either
absent or moderate inflammatory response are common
histopathologic findings in children with H. pylori infec-
tion (10,33). In 70% of our children with H. pylori in-
fection, bacteria density ranged from low to moderate.
We noted, however, that a higher degree of bacterial
colonization in the gastric mucosa was associated with
endoscopic nodular gastritis. In 1993, Eidt and Stolte
(34) studied an adult sample and noted that a higher
FIG. 3. Correlation between gastritis score found on histologic number of lymphoid follicles were observed in the antral
examination and endoscopic nodular gastritis. mucosa with the increase in H. pylori colonization den-

J Pediatr Gastroenterol Nutr, Vol. 36, No. 2, February 2003


sity. In our study, despite the low number of patients 3. Goodman KJ, Correa P, Aux HJT, et al. Helicobacter pylori in-
with endoscopic nodular gastritis associated with differ- fection in Colombian Andes: a population-based study of trans-
mission pathways. Am J Epidemiol 1996;144:290–9.
ent degrees of H. pylori density, which determined high 4. Mitchell HM, Li YY, Hu PJ, et al. Epidemiology of Helicobacter
CI in the statistical evaluation, our results still suggest pylori in Southern China: identification of early childhood as the
the presence of a strong association between both find- critical period for acquisition. J Infect Dis 1992;166:149–53.
ings. This finding was also recently demonstrated by 5. Staat MA, Kruszon-Moran D, McQuillan GM, et al. A population-
Luzza et al. (35). based serologic survey of Helicobacter pylori infection in children
and adolescents in the United States. J Infect Dis 1996;174:
Endoscopic nodular gastritis is not observed in all pa- 1120–3.
tients with H. pylori infection. Nodularity is probably 6. Gottrand F, Turck D, Vincent P. Helicobacter pylori infection in
related to the density of H. pylori at the beginning of early infancy. Lancet 1992;340:495.
infection, so that a large inoculum triggers an exagger- 7. Marshall BJ, Armstrong JA, McGechie DB, et al. Attempt to fulfill
ated immune response, as already suggested by Mitchell Koch’s postulates for pyloric Campylobacter. Med J Aust 1985;
et al. (13). The initial infection occurs more frequently 142:436–9.
8. Macarthur C, Saunders N, Feldman W. Helicobacter pylori, gas-
during childhood, and endoscopic nodular gastritis also troduodenal disease, and recurrent abdominal pain in children.
occurs more frequently in children than in adults. JAMA 1995;273:729–34.
H. pylori gastritis is usually less active and intense in 9. Belair PA, Metz DC, Faigel DO, et al. Receiver operator charac-
children than in adults (14). However, we demonstrated teristic analysis of endoscopy as a test for gastritis. Dig Dis Sci
that endoscopic nodular gastritis is not only associated 1997;42:2227–33.
with active chronic gastritis but also shows a significant 10. Bujanover Y, Reif S, Yahav J. Helicobacter pylori and peptic
disease in the pediatric patient. Pediatr Clin North Am 1996;43:
correlation with gastritis severity (Pearson correlation 213–34.
coefficient ⳱ 0.97). Studies with adult samples have 11. Genta RM, Hammer HW. The significance of lymphoid follicles in
investigated the relation between follicular gastritis and the interpretation of gastric biopsy specimens. Arch Pathol Lab
gastritis severity and have reported a significant correla- Med 1994;118:740–3.
tion between the presence of lymphoid follicles in the 12. Luzza F, Contaldo A, Imeneo M, et al. Testing for serum IgG
antibodies to Helicobacter pylori cytotoxin-associated protein de-
gastric mucosa and the grade and severity of gastritis tects children with higher grades of gastric inflammation. J Pediatr
(34,36,37). Gastroenterol Nutr 1999;29:302–7.
Luzza et al. (32,35) also observed a higher gastritis 13. Mitchell HM, Bohane TD, Tobias V, et al. Helicobacter pylori
score in children with antral nodularity independent of infection in children: potential clues to pathogenesis. J Pediatr
age. Those authors showed that gastritis score signifi- Gastroenterol Nutr 1993;16:120–5.
cantly increased with age, but they did not find a signifi- 14. De Giacomo C, Fiocca R, Villani L, et al. Helicobacter pylori
infection and chronic gastritis: clinical, serological, and histologic
cant age-related increase in antral nodularity. Although correlations in children treated with amoxicillin and colloidal bis-
we observed a significant increase in endoscopic nodular muth subcitrate. J Pediatr Gastroenterol Nutr 1990;11:310–6.
gastritis with increased age, we had few cases of nodu- 15. Prieto Bozano G, Polanco I, Larrauri J, et al. Helicobacter pylori
larity in younger children, which limited our statistical infection in children: clinical, endoscopic, and histologic correla-
conclusions. Therefore, this finding should be further tions. J Pediatr Gastroenterol Nutr 1992;14:420–5.
16. Rosh JR, Kurfist LA, Benkov KJ, et al. Helicobacter pylori and
investigated. The correlation of H. pylori density and gastric lymphonodular hyperplasia in children. Am J Gastroenterol
gastritis severity with increasing age was insignificant or 1992;87:135–9.
very low. These results, in association with those of the 17. Vandenplas Y, Blecker U, Devreker T, et al. Contribution of the
logistic regression model, suggest that the effects of H. 13
C-urea breath test to the detection of Helicobacter pylori gastritis
pylori density and increased gastritis score on gastric in children. Pediatrics 1992;90:608–11.
nodularity are independent of age. In addition, our study 18. Prieto Bozano G, Lorente Minarro M, Bejarano Lopez A, et al.
Antritis nodular y infeccion por Helicobacter pylori en el nino. An
suggests that the increase in gastritis score with gastric Esp Pediatr 1993;39:428–30.
nodularity depends on the increase in H. pylori density in 19. Elitsur Y, Raghuverra A, Sadat T, et al. Is gastric nodularity a sign
the gastric mucosa. for gastric inflammation associated with Helicobacter pylori infec-
Our study establishes the importance of antral nodu- tion in children? J Clin Gastroenterol 2000;30:286–8.
larity finding during endoscopic examination of children. 20. Dixon MF, Genta RM, Yardley JH, et al. Classification and grad-
ing of gastritis: the updated Sydney system. Am J Surg Pathol
This finding is associated with H. pylori infection and 1996;20:1161–81.
follicular gastritis, and it may identify patients in whom 21. McNulty CAM, Wise R. Rapid diagnosis of Campylobacter-
gastritis is more severe and associated with marked bac- associated gastritis. Lancet 1985;1:1443–4.
terial colonization independent of age. 22. Hassall E, Dimmick JE. Unique features of Helicobacter pylori
disease in children. Dig Dis Sci 1991;36:417–23.
23. Solari CA, Araruna RPN, Reis EMF, et al. Helicobacter pylori in
REFERENCES dyspeptic children and adults: endoscopic, bacteriologic, and his-
tologic correlations. Mem Inst Oswaldo Cruz 1994;89:581–6.
1. Oderda G. Transmission of Helicobacter pylori infection. Can J 24. Elitsur Y, Neace C. Detection of Helicobacter pylori organisms by
Gastroenterol 1999;13:595–7. Hp-Fast in children. Dig Dis Sci 1999;44:1169–72.
2. Mendall MA, Northfield TC. Transmission of Helicobacter pylori 25. Elitsur Y, Hill I, Lichtman SN, et al. Prospective comparison of
infection. Gut 1995;37:1–3. rapid urease tests (PyloriTek, CLO Test) for the diagnosis of Heli-

J Pediatr Gastroenterol Nutr, Vol. 36, No. 2, February 2003


cobacter pylori infection in symptomatic children: a pediatric mul- children infected with Helicobacter pylori with higher grades of
ticenter study. Am J Gastroenterol 1998;93:217–9. gastric inflammation. Gastrointest Endosc 2001;53:60–4.
26. Madani S, Rabah R, Tolia V. Diagnosis of Helicobacter pylori 33. Gottrand F, Cullu F, Turck D, et al. Normal gastric histology in
infection from antral biopsies in pediatric patients—is urease test Helicobacter pylori-infected children. J Pediatr Gastroenterol
that reliable? Dig Dis Sci 2000;45:1233–7. Nutr 1997;25:74–8.
27. Gold B, Colletti RB, Abbott M, et al. Helicobacter pylori infection 34. Eidt S, Stolte M. Prevalence of lymphoid follicles and aggregates
in children: recommendations for diagnosis and treatment. J Pe- in Helicobacter pylori gastritis in antral and body mucosa. J Clin
diatr Gastroenterol Nutr 2000;31:490–7. Pathol 1993;46:832–5.
28. Warren JR, Marshall B. Unidentified curved bacilli on gastric epi- 35. Luzza F, Pensabene L, Imeneo M, et al. Antral nodularity and
thelium in active chronic gastritis. Lancet 1983;1:1273–5. positive CagA serology are distinct and relevant markers of severe
29. Czinn SJ, Dahms BB, Jacobs GH, et al. Campylobacter-like or- gastric inflammation in children with Helicobacter pylori infection.
ganisms in association with symptomatic gastritis in children. Helicobacter 2002;7:46–52.
J Pediatr 1986;109:80–3. 36. Ladas SD, Rokkas T, Georgopoulos S, et al. Predictive factors and
30. Cadranel S, Goossens H, De Boeck M, et al. Campylobacter py- prevalence of follicular gastritis in adults with peptic ulcer and
loridis in children. Lancet 1986;1:735–6. nonulcer dyspepsia. Dig Dis Sci 1999;44:1156–60.
31. Lamireau T, Rigot A, Mégraud F, et al. Gastrite a Helicobacter 37. Zaitoun AM. The prevalence of lymphoid follicles in Helicobacter
pylori chez l’enfant. Arch Pediatr 1995;2:310–6. pylori associated gastritis in patients with ulcers and non-ulcer
32. Luzza F, Pensabene L, Imeneo M, et al. Antral nodularity identifies dyspepsia. J Clin Pathol 1995;48:325–9.

J Pediatr Gastroenterol Nutr, Vol. 36, No. 2, February 2003