H.pylori infection may mimic coeliac serology and morphology.


ESPGHAN 52. Annual Meeting, 5 - 08 Haziran 2019, sa.1


 Objectives and Study: Moderately elevated coeliac antibodies have been described in various

pathologies without any evidence of celiac disease (CD). Additionally, duodenal intraepithelial

lymphocytosis with a normal villous architecture (Marsh I lesion) is a relative common finding in

duodenal biopsies. The differential diagnosis includes bacterial overgrowth, NSAID injury in addition to

CD. H.pylori, which is the most common gastric pathogen, has been reported to induce morphological

changes in the duodenum ranging from intraepithelial lymphocytosis to crypt hyperplasia, as it is the

case in CD. Thus, the diagnosis of CD in the presence of H.pylori gastritis might be challenging. To

our knowledge, H.pylori associated coeliac seropositivity has not been considered before. In this

study, we analysed patients who underwent endoscopy because of a positive celiac serology, and

found to have concurrent H.pylori infection

 Methods: The endoscopy database were reviewed for the patients who underwent an endoscopy

because of elevated celiac serology, and 10 patients having concurrent H.pylori infection were

compiled. The demographic features, clinical manifestations, serologic tests and histopathology

records were reviewed. The histological features of gastritis were evaluated according to the modified

Sydney classification system. Duodenal histopathology was evaluated according the Marsh


 Results: The mean age was 8.3 ± 3.8 years and 8 of the patients were female. Of 10 patients 5 were

asymptomatic, and identified while routine screening for CD (2 type 1 DM, 1 Down syndrome, 1 Turner

syndrome and 1 hypothyroidism). All of the patients were anti-tTG positive and 3 out of 10 were also

positive for anti-EMA. Anti-tTG IgA level was higher than 5 times of the upper limit of normal in 5

patients, one of whom had concomitant anti-EMA positivity as well. Seven out of 10 patients had no

mucosal changes indicative of CD, One of the remaining 3 patients solely had intraepithelial

lymphocytosis that is consistent with Marsh 1, other 2 patients had partial villous atrophy in the distal

duodenum, and classified as Marsh 3A. All of the patients had H.pylori associative gastritis without any

sign of mucosal atrophy or intestinal metaplasia. All of the patients, infected with H.pylori received

eradication, thereafter checked for the serum anti-tTG IgA, and serum tTG titers decreased in all

patients, and 3 patients with positive EMA converted into negative as well. Two patients who had

Marsh 3A histopathology were evaluated by a repeat endoscopy, and mucosal recovery was detected

in both of them.

 Conclusion: Moderately elevated serum anti-tTG has been described in several autoimmune

diseases and infections (giardiasis). The association between H.pylori infection and elevated celiac

serology has not been reported before. Our results suggest that H.pylori infection may be related to

false positive celiac serology in individuals who have a predisposition to celiac disease. Given to the

fact that the prevalence of both H.pylori infection and celiac disease is high in some geographical

areas, a positive celiac serology should be carefully interpreted in the presence of an H.pylori