Celiac disease is frequently associated with chronic gastritis and immunologic disturbances (1). Because Helicobacter pylori is the causative agent in more than 90% of cases of chronic gastritis (2), a higher prevalence of this infection would be expected in patients with celiac disease. Furthermore, in patients with H. pylori infection the immune response to the bacterium may relate to disease outcome (2). The purposes of the current study were to assess the prevalence of H. pylori, the related symptoms, and the endoscopic and histologic gastric features in children with celiac disease.
MATERIALS AND METHODS
The study was approved by the local ethical committee. Eighty-one children (24 boys, 57 girls; age range, 1.4-17.7 years; median, 6.8 years) with celiac disease were studied. The diagnosis was based on the characteristic histologic finding of subtotal or severe partial villous atrophy and crypt hyperplasia in duodenal and jejunal biopsy specimens, and full remission while consuming a gluten-free diet. At the time of investigation, 49 children (12 boys, 37 girls; age range, 3.5-17.7 years; median 6.9 years) had been consuming a gluten-free diet for a median of 40 months (range, 3-192 months) and 32 (12 boys, 20 girls; age range, 1.4-16.4 years; median, 4.3 years) were untreated and had histologic subtotal (n = 10) or total (n = 22) villous atrophy.
All children had a blood sample taken at the time of investigation and stored at -20°C until tested. Serum H. pylori immunoglobulin (Ig) G was measured by means of a commercial kit (Helori-test IgG; Eurospital, Trieste, Italy) with 92% sensitivity and 91% specificity when assessed in a pediatric population (32 boys, 49 girls; age range, 2-18 years; median, 9 years) from the same geographic area, by using gastric biopsy specimens (histologic analysis and urease quick test) as reference (data not shown).
In 30 children (9 boys, 21 girls; age range, 1.4-17.2 years; median, 7 years) who underwent endoscopic procedures (gastroscopes GIF-XP20 ; Olympus Tokyo, Japan; and UGI-PE7; Fujinon, Tokyo, Japan) to obtain duodenal biopsy, at least three additional gastric biopsy specimens were taken: one from the antrum for urease quick test (Yamanouchi Pharma, Milan, Italy), one from the antrum, and one from the corpus for histologic examination. Informed parental consent was obtained for all endoscopic procedures performed.
Major symptom complaints were recorded at the time of appearance of celiac disease and evaluated according to H. pylori status. Furthermore, the presence of recurrent abdominal pain, defined according to Apley's criteria (3) as three or more episodes of localized or diffuse abdominal pain affecting the child's activities during a period of at least 3 months was recorded at the time of appearance of celiac disease and 3 months after a gluten-free diet was initiated. No child underwent anti-microbial therapy during this period.
Sections obtained in gastric biopsies were stained with hematoxylin and eosin and with Giemsa. The diagnosis of superficial chronic gastritis was based on the presence of inflammatory infiltration in the upper lamina propria. Interstitial gastritis, in which inflammation involved the full thickness of the mucosa, was distinguished from lymphocytic gastritis (LG), in which at least 25 lymphocytes per 100 gastric epithelial cells infiltrated the surface and pit epithelium. Furthermore, we considered an active inflammatory component to be present when neutrophils and eosinophils were seen in the lamina propria and regarded as lymphoid follicles any aggregation of lymphoid cells having a follicular structure with a germinative center. To clarify the nature of intraepithelial lymphocytes (IEL), additional paraffin sections were immunostained with T-cell (mouse antihuman T-cell monoclonal antibody, CD45RO, DAKO LSAB, Carpinteria, CA, U.S.A.) and B-cell (mouse antihuman B-cell monoclonal antibody, CD20, DAKO LSAB) markers.
Helicobacter pylori Status Determination
Children who underwent upper gastrointestinal endoscopy were classified as H. pylori positive when the urease quick test result was positive, when the organism was identified in the Giemsa-stained section in the antral or corpus gastric samples, or when both criteria were met. All the remaining children were categorized as positive or negative for H. pylori according to the results of serum H. pylori IgG assay.
Prevalence of Helicobacter pylori
Serum samples of 81 age- and sex-matched (one case, one control) children without celiac disease from the same geographic area attending the outpatient pediatric department for diagnostic investigations were used to compare the prevalence of H. pylori infection. A mixture of different clinical conditions having no proven influence on the occurrence of H. pylori infection were diagnosed (e.g., anorexia, nonorganic constipation, iron deficiency anemia, asthma, viral gastroenteritis, dermatitis, and cytomegalovirus infection).
Endoscopic and Histologic Gastric Features
The endoscopic and histologic gastric features of H. pylori-positive children with celiac disease were compared with those of H. pylori-negative children and with those of an H. pylori-positive age- and sex-matched control group without celiac disease.
Differences between groups were compared by the X2 test and Fisher's exact test, as appropriate. Odds ratio (OR) and 95% confidence intervals (CI) were also determined. P < 0.05 was required for significance.
Overall, 15 (18.5%) of 81 children with celiac disease and 14 (17.3%) of 81 control subjects were positive for H. pylori. The percentage of H. pylori positivity was similar in children with untreated (6 of 32; 18.7%) and treated (9 of 49; 18.4%) celiac disease. Signs and symptoms at the time of initial symptoms of celiac disease, according to H. pylori status, are summarized in Table 1. Recurrent abdominal pain was the only symptom that helped to distinguish between H. pylori-positive (6 of 15; 40%) and negative (5 of 66; 7.5%) children (p = 0.004; OR = 8.1; CI, 1.7-40). However, symptoms disappeared 3 months after gluten withdrawal, regardless of H. pylori status (Table 1). The gastric endoscopic and histologic findings are shown in Table 2. Frequency of antral nodularity was significantly (p = 0.033) higher in H. pylori-positive than H. pylori-negative children; and LG was a more common (p = 0.008) finding in H. pylori-positive children without celiac disease than in H. pylori-negative children without celiac disease. However, the frequency of these two findings did not differ between H. pylori-positive children with celiac disease and those without. All remaining endoscopic and histologic findings were similarly frequent, regardless of H. pylori status and of the presence of celiac disease. Immunohistochemical staining of gastric biopsy specimens showed that the IELs were almost exclusively T cells reactive to CD45RO monoclonal antibody. B-cell IELs marked with CD20 monoclonal antibody were usually absent.
In agreement with previous findings (4,5), this study shows that the prevalence of H. pylori infection in children affected by celiac disease is not higher than in control subjects. The increasing prevalence of H. pylori infection with increasing age (6) and the geographic variations in prevalence within the same country (7) makes the selection of control subjects important. In fact, in a resident population of the small rural town of Cirò in southern Italy, which lived in lower socioeconomic conditions than those in our community, a seroprevalence rate of H. pylori of 25% was found in children (median age of 7 years) (8). We carefully chose age- and sex-matched control subjects from the same geographic area who attended the outpatient pediatric department for diagnostic investigation, as did children with celiac disease. Furthermore, the prevalence of H. pylori infection in our children who had celiac disease did not change according to treatment, suggesting that the gluten-related improvement of immune abnormalities did not influence the acquisition of the infection. Possibly because of geographic and demographic differences, this finding is in contrast with that of Crabtree et al. (4) who showed a higher seroprevalence of H. pylori infection in untreated than in treated celiac patients.
Even though results in some studies have suggested that H. pylori infection may be more common among children with gastrointestinal symptoms, no specific symptomatology has been found to discriminate H. pylori-positive from H. pylori-negative subjects (9,10). With the exception of recurrent abdominal pain, no specific symptom was significantly associated with H. pylori infection in our celiac patients. However, that the symptoms disappeared after the withdrawal of gluten from the diet suggests that recurrent abdominal pain is not related to the presence of H. pylori.
As expected, the finding of antral nodularity was more common in H. pylori-positive than H. pylori-negative children (11). However, this feature cannot help to distinguish between H. pylori-positive children with and without celiac disease. Furthermore, the active component of the inflammatory response in H. pylori-positive children was found to be lower than that reported in adults (11). Lymphocytic gastritis is a histologic entity characterized by a marked increase in the number of IELs in the gastric mucosa. Immunohistochemical and morphometric studies have shown that IELs are T cells and that their number is significantly higher than in non-lymphocytic gastritis and in normal gastric specimens (5,12). Little is known of the cause or natural history, but the association with H. pylori and celiac disease has been documented (5,12). Possibly because of selection bias (e.g., immunologically predisposed people, subjects with treated or untreated celiac disease, diagnostic test for H. pylori, subject's age) and small sample size, an apparent predominance of LG in H. pylori-positive children (3 of 7 vs. 1 of 30 children with celiac disease) was shown. Furthermore, it is noteworthy that the number of IELs in many patients with celiac disease increased (e.g., 13-18 per 100 epithelial cells) but did not reach the cutoff level for diagnosis of LG. Nevertheless, LG was similarly frequent in H. pylori-positive children with and without celiac disease. Mucosa-associated lymphoid tissue resembling follicles can develop in the gastric mucosa as a result of interaction between different stimuli and specific HLA haplotypes, as in celiac disease (13), or after H. pylori infection (14). In this study, lymphoid follicles were observed in a similar proportion in children with and without celiac disease who were H. pylori positive, suggesting that interaction between H. pylori and celiac disease do not potentiate their development.
In conclusion, our results show that H. pylori infection is not increased in children with celiac disease, and that the clinicopathologic pattern of the infection is not specifically influenced in this condition.
1. Gilberg R, Kastrup W, Mobacken H, Stockbrugger R, Ahren C. Gastric morphology and function in dermatitis herpetiformis and in coeliac disease. Scand J Gastroenterol
2. Blaser MJ. Hypotheses on the pathogenesis and natural history of Helicobacter pylori
-induced inflammation. Gastroenterology
3. Apley J. The Child With Abdominal Pains
. 2nd ed. Oxford, UK: Blackwell Scientific, 1975.
4. Crabtree JE, Mahony SO, Wyatt JI, et al. Helicobacter pylori
serology in patients with coeliac disease and dermatitis herpetiformis. J Clin Pathol
5. De Giacomo C, Gianatti A, Negrini R, et al. Lymphocytic gastritis
: A positive relationship with celiac disease
. J Pediatr
6. Parsonnet J. The incidence of Helicobacter pylori
infection. Aliment Pharmacol Ther
7. Palli D, Vaira D, Menegatti M, Saieva C. A serologic survey of Helicobacter pylori
infection in 3281 Italian patients endoscoped for upper gastrointestinal symptoms. Aliment Pharmacol Ther
8. Luzza F, Imeneo M, Maletta M, et al. Seroepidemiology of Helicobacter pylori
infection and hepatitis A in a rural area: Evidence against a common mode of transmission. Gut
9. Gormally SM, Drumm B. Helicobacter pylori
and gastrointestinal symptoms. Arch Dis Child
10. Ashorn M. What are the specific features of Helicobacter pylori gastritis
in children? Ann Med
11. Mitchell HM, Bohane TD, Tobias V, et al. Helicobacter pylori
infection in children: Potential clues to pathogenesis. J Pediatr Gastroenterol Nutr
12. Dixon M, Wyatt J, Burke D, Rathbone B. Lymphocytic gastritis
-relationship to Campylobacter pylori
infection. J Pathol
13. Cammarota G, Fedeli G, Tursi A, Corazza GR, Gasbarrini G. Coeliac disease and follicular gastritis
14. Wotherspoon AC, Ortiz-Hidalgo C, Falzon MR, Isaacson PG. Helicobacter pylori
and primary B-cell gastric lymphoma. Lancet
Keywords:© 1999 Lippincott Williams & Wilkins, Inc.
Helicobacter pylori; Celiac disease; Gastritis; Lymphocytic gastritis; Lymphoid follicles; Recurrent abdominal pain