Celiac disease (CD) is a chronic autoimmune enteropathy caused by the ingestion of gluten, a component of wheat, barley, and rye, which affects genetically susceptible individuals and is characterized by typical lesions of the duodenal mucosa. Recent studies revealed that CD affects >1% of the general population, both in Europe (1,2) and in North America (3).
CD can appear with gastrointestinal symptoms (typical form), in patients with iron-deficiency anemia, dermatitis herpetiformis, headaches, recurrent aphthous stomatitis (atypical form), or be asymptomatic (silent form). Several long-term complications, such as autoimmune disorders (4,5), osteoporosis (6), infertility (7,8), malignancy (9), or a refractory CD, have also been described.
Anti-gliadin (AGA), anti-endomysium (EMA), and anti-human transglutaminase antibodies (tTG Ab) can be used as a screening method. The demonstration of histological changes of the small bowel mucosa, evaluated according to the Marsh classification as modified by Oberhuber et al (10), is the criterion standard for CD diagnosis.
In CD, an involvement of the gastric mucosa was also described (11–16), particularly lymphocytic gastritis (LG), a form of gastritis of uncertain pathogenesis, reported for the first time by Haot et al (17). According to Sydney Updated System (18), LG was ranked among the special forms of gastritis and is histologically characterized by mature lymphocytes infiltrating the surface and foveolar epithelium (>25 lymphocytes/100 cells), in conjunction with a variable increase in chronic inflammatory cells in the lamina propria of the gastric mucosa. It has been shown that the number of intraepithelial lymphocytes (IEL) in the gastric mucosa of celiac patients is significantly higher than in controls, regardless of the diagnosis of LG and, that this number can decrease in patients with gluten-free diet (11).
The aim of our study was to perform a global evaluation of gastric mucosa in a large series of children and adolescents with CD and in gastroenterological controls (GC).
In this retrospective study, a total of 380 patients (148 boys, age range 11 months–19 years, median age 6.3 years), who underwent upper endoscopy for several gastrointestinal complaints, were evaluated. These patients were divided into 2 groups according to the diagnosis:
1. A total of 226 celiac children and adolescents (82 boys, age range 11 months–19 years, median age 5.7 years) at gluten-containing diet
2. A total of 154 children with GC (66 boys, age range 1–18 years, median age 7.4 years) who underwent endoscopy because of poor growth, chronic abdominal pain, gastroesophageal reflux, but were negative for both CD-specific autoantibodies and histological duodenal lesions
For each subject, a form containing data about anthropometric parameters, information about clinical evaluation, and serological tests was completed to classify patients in typical form (gastrointestinal symptoms), atypical form (iron-deficiency anemia, dermatitis herpetiformis, headaches, recurrent aphthous stomatitis), or silent form if asymptomatic.
1. Total serum IgA
2. AGA IgA and IgG autoantibodies (only for children younger than 3 years) by enzyme-linked immunosorbent assay (ELISA), using α-Kit Eurospital (Trieste, Italy)
3. EMA IgA and IgG by indirect immunofluorescence (Eurospital, Trieste, Italy)
4. Anti-transglutaminase IgA and IgG by enzyme immunoassay (ELISA Kit Eurospital, Trieste, Italy)
5. Gastric parietal cell antibodies (GPCA) by ELISA (AXA Diagnostics, S.r.l., Rome, Italy).
After narcosis, the esophagogastroduodenoscopy was performed with Olympus PQ-20, GIF-E, or P140 endoscopes, on the patients who were fasting overnight.
During endoscopy, multiple biopsies were taken with forceps FB-19k-1 and FB-24Q (2 fragments from the duodenal bulb, 4 from the distal duodenum, and 4 from the gastric antrum—2 for histology and 2 for the rapid urease test). Each biopsy was oriented on Millipore paper, fixed in 4% formalin, and embedded in paraffin. The sections, stained with hematoxylin-eosin and Giemsa, were examined with a light microscopy.
Gastric lesions were classified according to the Updated Sydney System that encompasses several groups of nosological entities, for convenience divided into 3 broad categories: acute, chronic, and special (or “distinctive”) forms. Gastric inflammation was classified as follows: grade 1 for mild lesions, grade 2 for moderate lesions, and grade 3 for severe lesions (18).
Helicobacter pylori (Hp) infection was diagnosed by rapid urease testing and by histology.
The histological lesions of the intestinal mucosa were defined according to Marsh classification modified by Oberhuber et al, which distinguishes:
Type 0: Normal mucosa
Type 1: Infiltrative pattern (>40 IEL/100 enterocytes)
Type 2: Hyperplastic pattern (increased IEL and crypt hyperplasia)
Type 3: Villous atrophy (3a: mild; 3b: subtotal; 3c: total) (10)
The diagnosis of CD was performed according to the North American Society for Gastroenterology, Hepatology, and Nutrition criteria (19). The study was conducted in accordance with the principles of the Declaration of Helsinki.
The statistical analysis of the results was performed with SPSS version 16.0 (SPSS Inc, Chicago, IL). The comparison between the percentages was performed with the χ2 test, and the type of correlation was determined by the Pearson correlation coefficient. Two-tailed P values <0.05 were considered significant.
Of the 226 celiac children (group 1), 137 (61%) showed a typical, 33 (15%) an atypical, and 56 (24%) a silent clinical form. The most frequent symptoms were abdominal pain, diarrhea, abdominal distension, and poor weight gain. The ratio of typical forms and the sum of atypical and silent (subclinical) forms was 1.5:1.
The AGA, determined in all of the 55 children younger than 3 years, were positive in 50 (91%). IgA EMA were positive in 200 children (88%); IgA tTG Ab were positive in 211 patients (93%). The only girl with IgA deficiency was positive for IgG tTG Ab and IgG EMA.
A total of 31 children (13.7%) showed macroscopic abnormalities localized in the stomach, such as 12 erosions, mainly located at the bottom, 5 petechiae at the bottom, antrum, and corpus, 6 antrum nodularity, 4 hyperemia, and 4 paleness of mucosa.
Histological lesions of gastric mucosa were present in 65 patients (28.7%). In particular, chronic superficial gastritis (CSG) (Fig. 1) was found in 48 patients (21.2%), LG (Fig. 2) in 16 (7%), and interstitial gastritis (IG) in 1 child (0.5%) (Table 1) (10). The LG was found more frequently in older patients (P < 0.046).
The lesions were present with varying degrees of severity (Table 2). Moderate (grade 2) and severe (grade 3) grades were more frequent in subjects with LG (P < 0.01).
Ten children with CSG (20.8%) showed macroscopic alterations of the gastric mucosa: hyperemia, petechiae, erosions, and nodularity of the antrum. Only 1 patient with LG (6.2%) showed the presence of erosions of the gastric mucosa.
Hp infection was diagnosed in 6 patients: 4 with CSG (8.3%), 1 with an LG (6.2%), and 1 patient with IG. Hp infection correlated with macroscopic alterations in the stomach (P < 0.01). Hp infection was more frequent in subjects with more severe degrees of gastritis (P < 0.01). Table 1 shows CD clinical forms in patients divided according to the aspect of gastric mucosa.
In the duodenum of 56 patients (24.7%), scalloped folds were observed, in 51 (22.6%) flattening of folds, and in 16 (7%) an edematous appearance. In addition, 18 children (8%) had nodularity of the duodenal mucosa, 11 (4.9%) showed a pale mucosa, 9 (4%) a friable mucosa, and 3 (1.3%) mucosa with a mosaic appearance. A total of 99 patients (44%) had a normal duodenal endoscopic appearance.
At histological examination, 219 children (97%) showed villous atrophy (87% of them type 3c, 11% 3b, and 2% 3a), 4 (1.7%) hyperplasic crypts (type 2), and 3 (1.3%) showed only an increase (>40/100 epithelial cells) of IEL (type 1). In patients with type 1 or 2 duodenal lesions, the diagnosis was supported by the presence of CD-specific autoantibodies. In 210 children (93%), the typical histological lesions of CD had a diffuse distribution, in 13 (6%) children lesions were localized only in the bulb, and 3 (1%) children had a patchy distribution.
The most severe histological changes of CD (type 3c) were present in all of the children with LG, in the child with IG, and in 42 (87%) patients with CSG.
In group 2, consisting of 154 GC, the main reported symptoms were chronic abdominal pain, abdominal distension, alterations of growth, and poor weight gain. In all of these patients, CD-specific antibodies as well as duodenal histological findings were negative.
At endoscopic examination, 3 children had hiatal hernia, 3 had cardial incontinence, whereas 34 children (22%) showed alterations of gastric mucosa: in 19 nodularity (mainly located in the antrum), in 8 petechiae (in the corpus and the antrum) and 1 friable mucosa.
At histological examination, chronic superficial gastritis was found in 33 patients (21.4%) (21 with Hp infection) (Table 1). No control subject showed LG. A moderate (grade 2) intestinal metaplasia was also present in a child with gastritis. Overall, Hp infection was found in 8.3% of celiac patients and in 64% of controls with CSG (P < 0.0001).
It was possible to determine GPCA in 90 celiac children (36 with CSG, 11 with LG, and 43 patients with normal gastric mucosa). Positive values were found in 10 children (2 [18.2%] with LG, 4 [11%] with CSG, and 4 [9.3%] with normal gastric mucosa). These autoantibodies were prevalent in children with macroscopic alterations of gastric mucosa (P < 0.05).
The GPCA were evaluated in 29 controls (3 with CSG). Only 1 child (3.4% [with normal gastric mucosa]) was positive for these antibodies (Table 1).
In the present study, we demonstrated a gastric involvement with varying degrees of severity in about one-third of a large number of children and adolescents at the diagnosis of CD, receiving a gluten-containing diet: 48 patients (21.2%) had chronic superficial gastritis, 16 (7%) had LG, and only 1 (0.5%) had interstitial gastritis.
The gastric involvement, therefore, in CD is frequent and confirms the theory that gluten intolerance may involve other portions of the gastrointestinal tract, as also demonstrated in oral (20) and colonic mucosa (21). The chronic superficial gastritis was the most frequent form, which was present in 21.2% of celiacs and in 21.4% of controls. The majority of celiacs with CSG (79%) had a symptomatic form of CD (typical plus atypical). Our data suggest that in CD, as well as in the controls, a CSG can coexist with Hp infection. In approximately one-fifth of patients with CSG, macroscopic alterations of the gastric mucosa were detected and the presence of Hp infection correlated with the presence of macroscopic alterations. The lower prevalence of Hp infection in celiacs with CSG (8.3%), with respect to the GC with CSG (64%), allows us to assume that the pathogenesis of gastric damage in CD may be affected by other factors, possibly including the delay of gastric emptying, which is reported in 30% to 60% of celiacs (22). Our data are in agreement with the study of Oderda et al (23), showing a higher number of CSG in patients with CD receiving gluten-containing diet (43.5%) with respect to controls (28.2%). Moreover, authors demonstrate that CSG in celiacs receiving gluten-free diet is less frequent (29.4%), thus suggesting a role of gluten in CSG pathogenesis.
In only 1 study on children (12) and in the study by Vogelsang et al on adults (13), the prevalence of LG was higher (42% and 84%, respectively) than in our series; however, Vogelsang et al (13) have chosen a different cutoff (>8 lymphocytes/100 cells), whereas we followed the Sydney System classification (>25 lymphocytes//100 cells). This finding agrees with the demonstration of a positive correlation between the counts of IEL in gastric mucosa and severity of the damage of the duodenal mucosa (11). All of the patients with LG in our study had total villous atrophy (type 3c), and the percentage of patients with LG with diffuse damage was significantly greater than in other groups of celiacs. In addition, among the 16 patients with LG, 9 (56%) had a subclinical form (2 atypical and 7 silent) and the mean age was higher than in subjects with CSG or without gastritis. It is to be presumed, therefore, that LG is found easily in patients with CD who mostly come late to the diagnosis, perhaps because they are mildly symptomatic, and in ones who have a longer duration of illness and exposure to gluten, resulting in increased severity and extent of histological damage in the duodenum, as well as involvement of other sections of the digestive tract. This could also explain the lower prevalence of LG in children with CD than in adults.
A possible underestimation of the prevalence of LG in our patients may be the consequence of the biopsies taken only from the gastric antrum; however, Hayat et al (14) have found LG-typical lesions predominantly in the gastric antrum of celiac patients.
The role of Hp infection in the pathogenesis of LG is debated. In the series of Prasad et al (12), the prevalence of Hp in patients with CD and LG was found to be similar to ours (6%). A much higher percentage (57%) was found in a group of 70 adult celiac patients with LG and a neutrophilic gastritis (15). It was also noted that Hp infection may play a role in the pathogenesis of LG also in nonceliac subjects, and the eradication of the bacterium may lead to a significant reduction in the number of IEL, a reduction of inflammation of the body, and an improvement of dyspeptic symptoms (14). Other authors, however, found no significant differences in counts of IEL between celiac patients with or without Hp infection or between Hp-positive and Hp-negative controls (16). In our study, 6 patients with CD (2.7%) were Hp positive and Hp infection was found more frequently in subjects with more severe degrees of gastritis. Among the patients with LG, however, only 1 (6.2%) was Hp positive. Nevertheless, in our control group, the prevalence of Hp was 15.6%, but none of them showed an LG. Our data, therefore, lead us to hypothesize that the role of the bacterium in the pathogenesis of LG is marginal. In our series, primary CSG was found in 7.8 % of GC, similar to the French series (24).
Macroscopically, just 1 of those with LG (6.2%) showed the presence of erosions of the gastric mucosa. This demonstrates that the LG, which was once considered only in patients with macroscopic lesions (“varioliform appearance”) (14), may also occur in patients with endoscopically normal gastric mucosa.
A novelty of our study is the determination of GPCA in celiac children and adolescents and in the gastroenterological control group, and the search of a correlation with histological lesions. The higher percentage of positivity was found in celiacs with LG (18.2%), followed by those with CSG (11%) and celiacs with normal gastric mucosa (9.3%). These antibodies were present in only 3.5% of controls.
The GPCA are present in 90% of patients with pernicious anemia, but they are also found in the serum of patients with other autoimmune disorders such as autoimmune thyroiditis (25), primary biliary cirrhosis (26), and diabetes mellitus type 1 (27). In the latter case, the level of GPCA increases with age and duration of disease (27). Their presence, therefore, may represent an early marker of gastric atrophy (28), and the level of these antibodies correlates with the degree of atrophy in the gastric body. In a study on a large series of healthy adult volunteers, the prevalence of gastric atrophy was 3.5% (29), whereas in healthy children the prevalence is unknown (30). To date, GPCA have been reported in 3.6% of celiac adults (31). In our study, the presence of these antibodies does not correlate with hemoglobin values or the values of mean corpuscular volume (data not shown), not establishing a precise clinical consideration. It could therefore be interpreted in the context of a multiorgan autoimmune process that is characteristic of CD, as confirmed by the higher positivity in LG celiac patients.
In conclusion, this is the first study to our knowledge to systematically evaluate the gastric mucosa involvement in a large series of children and adolescents with CD and in GC. Our results confirm that CD appears once again as a multiform entity that may involve many organs other than the duodenum in terms of both histological damage and autoimmune process.
The authors thank Mrs Patricia Byrne for helping with English.
1. Maki M, Mustalahti K, Kokkonen J, et al. Prevalence of coeliac disease among children in Finland. N Engl J Med 2003; 348:2517–2524.
2. Bonamico M, Nenna R, Montuori M, et al. First salivary screening of celiac disease by detection of antitransglutaminase autoantibody radioimmunoassay in 5000 Italian primary school children. J Pediatr Gastroenterol Nutr 2011; 52:17–20.
3. Catassi C, Fasano A. Celiac disease. Curr Opin Gastroenterol 2008; 24:687–691.
4. Ansaldi N, Palmas T, Corrias A, et al. Autoimmune thyroid disease and celiac disease in children. J Pediatr Gastroenterol Nutr 2003; 37:63–66.
5. Salardi S, Volta U, Zucchini S, et al. Prevalence of celiac disease in children with type 1 diabetes mellitus increased in the mid-1990 s: an 18-year longitudinal study based on anti-endomysial antibodies. J Pediatr Gastroenterol Nutr 2008; 46:612–614.
6. Stazi AV, Trecca A, Trinti B. Osteoporosis in celiac disease and in endocrine and reproductive disorders. World J Gastroenterol 2008; 4:498–505.
7. Sher KS, Mayberry JF. Female fertility, obstetric and gynaecological history in coeliac disease: a case control study. Acta Paediatr 1996; 412:76–77.
8. Nenna R, Mennini M, Petrarca L, et al. Immediate effect on fertility of a gluten-free diet in women with untreated coeliac disease. Gut 2011; 60:1023–1024.
9. Mathus-Vliegen EM, Van Halteren H, Tytgat GN. Malignant lymphoma in celiac disease: various manifestations with distinct symptomatology and prognosis. J Int Med 1994; 236:43–49.
10. Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol 1999; 11:1185–1194.
11. Alsaigh N, Odze R, Goldman H, et al. Gastric and esophageal intraepithelial lymphocytes in pediatric coeliac disease. Am J Surg Pathol 1996; 20:865–870.
12. Prasad KK, Thapa BR, Lal S, et al. Lymphocytic gastritis and Indian childhood celiac disease: evidence of positive relationship. J Pediatr Gastroenterol Nutr 2008; 47:568–572.
13. Vogelsang H, Oberhuber G, Wyatt J. Lymphocytic gastritis and gastric permeability in patients with celiac disease. Gastroenterology 1996; 111:73–77.
14. Hayat M, Arora DS, Wyatt JI, et al. The pattern of involvement of the gastric mucosa in lymphocytic gastritis is predictive of the presence of duodenal pathology. J Clin Pathol 1999; 52:815–819.
15. Feeley KM, Heneghan MA, Stevens FM, et al. Lymphocytic gastritis and coeliac disease: evidence of a positive association. J Clin Pathol 1998; 51:207–210.
16. Karttunen T, Niemela S. Lymphocytic gastritis and coeliac disease. J Clin Pathol 1990; 43:436–437.
17. Haot J, Hamichi L, Wallez L, et al. Lymphocytic gastritis: a newly described entity: a retrospective endoscopic and histological study. Gut 1988; 29:1258–1264.
18. Dixon MF, Genta RM, Yardley JH, et al. Classification and grading of gastritis: the Updated Sydney System. International workshop on the histopathology of gastritis, Houston 1994. Am J Surg Pathol 1996; 20:1161–1181.
19. Husby S, Koletzko S, Korponay-Szabó IR, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr 2012; 54:136–160.
20. Pastore L, Carroccio A, Compilato D, et al. Oral manifestations of celiac disease. J Clin Gastroenterol 2008; 42:224–232.
21. Casella G, Villanacci V, Di Bella C, et al. Colonoscopic findings in coeliac disease on a gluten free diet. Rev Esp Enferm Dig 2010; 102:538–541.
22. Rashtak S, Murray JA. Persistent nausea and abdominal pain in a patient with delayed gastric emptying: shall we think of celiac disease? Clin Gastroenterol Hepatol 2009; 7:910.
23. Oderda G, Forni M, Morra I, et al. Endoscopic and histologic findings in the upper gastrointestinal tract of children with coeliac disease. J Pediatr Gastroenterol Nutr 1993; 16:172–177.
24. Kalach N, Papadopoulos S, Asmar E. In French children, primary gastritis is more frequent than Helicobacter pylori gastritis. Dig Dis Sci 2009; 54:1958–1965.
25. García García B, Gimeno Orna JA, Aguillo Gutiérrez E, et al. Prevalence and predictive factors of parietal cell antibody positivity in autoimmune thyroid disease. Endocrinol Nutr 2010; 57:49–53.
26. Ciesek S, Becker T, Manns MP, et al. Anti-parietal cell autoantibodies (PCA) in primary biliary cirrhosis: a putative marker for recurrence after orthotopic liver transplantation? Ann Hepatol 2010; 9:181–185.
27. Karavanaki K, Kakleas K, Paschali E, et al. Screening for associated autoimmunity in children and adolescents with type 1 diabetes mellitus (T1DM). Horm Res 2009; 71:201–206.
28. Lo CC, Hsu PI, Lo GH, et al. Implications of anti-parietal cell antibodies and anti-Helicobacter pylori antibodies in histological gastritis and patient outcome. World J Gastroenterol 2005; 11:4715–4720.
29. Telaranta-Keerie A, Kara R, Paloheimo L, et al. Prevalence of undiagnosed advanced atrophic corpus gastritis in Finland: an observational study among 4,256 volunteers without specific complaints. Scand J Gastroenterol 2010; 45:1036–1041.
30. Dimitrov G, Gottrand F. Does gastric atrophy exist in children? World J Gastroenterol 2006; 12:6274–6279.
31. da Rosa Utiyama SR, da Silva Kotze LM, Nisihara RM, et al. Spectrum of autoantibodies in celiac patients and relatives. Dig Dis Sci 2001; 46:2624–2630.