THE RADIAL DISTRIBUTION OF DILATED INTERCELLULAR SPACES OF THE ESOPHAGEAL SQUAMOUS EPITHELIUM IN PATIENTS WITH REFLUX DISEASE EXHIBITING DISCRETE ENDOSCOPIC LESIONS.

INTRODUCTION: Dilatation of intercellular spaces of the esophageal squamous epithelium has been suggested as a marker of early acid reflux-induced damage. This change is a potentially useful addition to histomorphological changes that represent so called minimal endoscopic lesions. We have assessed dilatation of intercellular spaces with regard to: (1) interobserver variability, and (2) whether the incidence of this varies between 'red streaks' and the adjacent normal looking squamous epithelium. METHODS: Esophageal biopsies from 44 patients with chronic gastro-esophageal reflux (GERD) were evaluated. At endoscopy, these patients had one or more red streaks on the tops of the mucosal folds in the distal esophagus. Biopsies were taken from the red streaks and from the normal-appearing mucosa 1 cm lateral to the red streaks. Biopsies were assessed in a blinded fashion by two independent pathologists (MV & RF). Criteria for assessing intercellular space dilatation were evaluated and agreed on prior to the study. RESULTS: Good interobserver agreement was recorded (kappa = 0.82 at the streaks and 0.77 for the control tissues) for absence/presence of intercellular space dilatation. Red streak and control biopsies differed significantly (p = 0.0001), with respect to presence of dilated intercellular spaces, with 90.5 % of the former demonstrating this as present compared to 56.1% in the controls. CONCLUSION: This study supports the concept that esophageal mucosal minimal changes due to reflux is localised and that dilatation of intercellular spaces is an early sign of reflux-induced epithelial damage. The low interobserver variability in the assessment of intercellular space dilatation suggests that this may be a useful variable for assessment of early signs of acid-reflux induced damage to the squamous epithelium of the esophagus by use of light microscopy.