Barrett's oesophagus and metaplasia at the oesophagogastric junction : an epidemiological approach

Abstract

The aims of this thesis were to estimate the prevalence of Barrett's oesophagus (B0) and related histological aberrations, to investigate agreement between endoscopy and histology regarding the diagnosis of BO and to investigate risk factors for BO paying particular attention to epidemiological aspects of the study design. We also investigated pancreatic acinar metaplasia (PAM) with special emphasis in its relation to gastro-oesophageal reflux and BO.

Seven hundred and sixty-nine patients endoscoped for the first time at endoscopy units exclusively serving defined catchment areas in southeast Sweden were examined and clinical data recorded. Exposure data were also collected from 160 population controls. In a subsample of 26 patients 24 hour oesophageal pH monitoring were performed.

Overall intestinal (M prevalence in the distal oesophagus and/or gastric cardia was 14%. BO was noted in 4%, with a predominance of women (69%) Both prevalence of IM overall and cardia type mucosa in the gastric cardia were significantly associated with increasing age.

Overall concordance between endoscopy and histology regarding columnar mucosa above the oesophagogastric junction was 74% (95% confidence interval [Cl] 71-77%) and the agreement beyond chance was fair (Kappa=0.38, 95% Cl 0.32-0.45). Our data were consistent with a lower threshold for macroscopic detection of columnar epithelium above the oesophagogastric junction, when risk factors for BO were present.

Reflux symptoms and smoking indicated 10.7- and 3.3-fold risks, respectively, for BO (95% Cl 3.5-33.4 and 1.1-9.9 respectively) in the comparison with population controls. Body mass was unrelated to risk. In the cross-sectional analysis among endoscopy room patients, reflux symptoms were associated with an odds ratio (OR) of 2.0 (95% Cl 0.8-5.0) This association was, however, modified by the subjunctional presence of Helicobacterpylori; although the infection was not in itself importantly connected with risk, a combination of reflux symptoms and H. pylori was linked to an almost five-fold risk (95% Cl 1.4-16.5), as compared with the absence of both factors.

PAM was found above the oesophagogastric junction (OGJ) in 9% and below the OGJ in 13% (2% had PAM both above and below the OGJ PAM below and PAM exclusively above the OGJ were both borderline associated with age, with a 2% increase in prevalence per year. PAM exclusively above the OGJ was significantly associated with female gender (OR 2.8, 95% Cl 1.3-6.2) and subjunctional presence of H. pylori (OR 2.3, 95% Cl 1.1-4.9). Among patients with BO 38% had PAM above the OGJ Mean values for percentage time with oesophageal pH<4.0 indicated pathological gastro-oesophageal reflux among patients who had PAM without and with accompanying BO.

In conclusion, while BO is not very common among Swedish gastroscopy patients, IM and PAM are found in every 7th and 5th patient, respectively. Age-dependent increments in prevalence suggest that not only BO and IM, but also cardia type mucosa, are acquired and/or progressive lesions. The agreement between macroscopic and microscopic assessments of BO is no more than fair, and partly dependent on the presence of patient characteristics suggestive of pathology in the region. Reflux is the dominating risk factor for BO, and proximal gastric colonization of H. pylori seems to amplify this risk. PAM might be an age-dependent lesion, associated with H. pylori, female gender and gastro-oesophageal reflux if located above the OGJ.