Extent of Barrett's metaplasia: a prospective study of the serial change in area of Barrett's measured by quantitative endoscopic imaging

Abstract

Background: An accurate determination of the extent of Barrett's metaplasia is critical to the study of its natural history and response to therapy. Our hypothesis is that area calculations offer advantages over length estimates of Barrett's. Methods: Changes in both measures and estimates of progression or regression between two endoscopies in 17 patients were compared. Area was calculated using a computer image analysis technique. Results: Although there was no significant difference in length correlation versus area correlation between endoscopies ( r = 0.90 vs 0.99), the mean change in absolute length (1.4 ± 0.2 cm) was greater than the change in area (4.5 ± 1.4 cm 2, equivalent to a length of 0.67 ± 0.2 cm, p = 0.001). The percent change in absolute length (26.9%) was greater than the change in area (16%, p = 0.001). Discordance of estimates of progression or regression between area and length was found in nine patients. The image technique detected no change in the area of squamous islands. Conclusions: Image analysis can precisely measure the extent of Barrett's, including squamous islands. Area showed little change, whereas measures of length were more varied. Computer based image analysis provides a more precise estimate of interval change of Barrett's. (Gastrointest Endosc 1997;45:456-62.)