OBJECTIVES

Recent studies indicate that colonization with cagA-positive Helicobacter pylori (H. pylori) strains may protect against gastroesophageal reflux disease (GERD) and its complications, but the role of cagA in the etiology of Barrett's esophagus has so far been poorly investigated. The pathogenesis of intestinal metaplasia (IM) at an endoscopically normal esophagogastric junction (EGJ) is still unclear, and the role of the H. pylori virulence factor cagA in it has not been investigated. The aim of our study was to assess the relationship between H. pylori and cagA-positive H. pylori in particular and IM at an endoscopically normal EGJ and Barrett's esophagus.

METHODS

Serum samples were obtained from 62 patients without IM, 43 patients with IM at an endoscopically normal junction, and 51 patients with Barrett's esophagus. IM was defined as presence of goblet cells with positive staining with Alcian blue. The prevalence of H. pylori and cagA was investigated by assessment of IgG antibody levels as determined by ELISA.

RESULTS

The overall H. pylori prevalence was 59% (92/156), and the cagA prevalence was 29% (46/156). Although 63% (39/62) of IM negative subjects and 74% (32/43) of those with IM at the junction were H. pylori positive, only 41% (21/51) of Barrett's patients tested positive. The differences between the IM negative and the Barrett's group (p = 0.02) and between IM at the junction and Barrett's were significant (p = 0.002). The relative cagA prevalence (percentage with cagA positivity and H. pylori positivity) was 56% (22/39) in patients who were IM negative, 59% (19/32) in those with IM at the junction, and 24% (5/21) in those with Barrett's. The prevalence of anti-CagA was significantly lower in patients with Barrett's esophagus compared with patients who were IM negative (p = 0.002) and those who had IM at the junction (p < 0.001). No difference in cagA prevalence was seen between the latter groups.

CONCLUSIONS

These findings are in line with the concept that H. pylori and cagA-positive strains in particular protect against the development of Barrett's esophagus. In contrast, our findings do not support the theory that IM at an endoscopically normal esophagogastric junction is associated with H. pylori or cagA-positive strains. IM at the junction and Barrett's esophagus seem to have different etiologies.