Effect of segment length on risk for neoplastic progression in patients with Barrett esophagus

Background: The increased risk for esophageal adenocarcinoma associated with long-segment (greater than or equal to 3 cm) Barrett esophagus is well recognized. Recent studies suggest that short-segment (<3 cm) Barrett esophagus is substantially more common; however, the risk for neoplastic progression in patients with this disorder is largely unknown. Objective: To examine the relation between segment length and risk for aneuploidy and esophageal adenocarcinoma in patients with Barrett esophagus. Design: Prospective cohort study. Setting: University medical center in Seattle, Washington. Patients: 309 patients with Barrett esophagus. Measurements: Patients were monitored for progression to aneuploidy and adenocarcinoma by repeated endoscopy with biopsy for an average of 3.8 years. Cox proportional hazards analysis was used to calculate adjusted relative risks and 95% Cls. Results: After adjustment for histologic diagnosis at study entry, segment length was not related to risk for cancer in the full cohort (P > 0.2 for trend). When patients with high-grade dysplasia at baseline were excluded, however, a nonsignificant trend was observed; based on a linear model, a 5-cm difference in segment length was associated with a 1.7-fold (95% CI, 0.8-fold to 3.8-fold) increase in cancer risk. Among all eligible patients, a 5-cm difference in segment length was associated with a small increase in the risk for aneuploidy (relative risk, 1.4 [CI, 1.0 to 2.1]; P = 0.06 far trend). A similar trend was observed among patients without high-grade dysplasia at baseline. Conclusions: The risk for esophageal adenocarcinoma in patients with short-segment Barrett esophagus was not substantially lower than that in patients with longer segments. Although our results suggest a small increase in risk for neoplastic progression with increasing segment length, additional follow-up is needed to determine whether the patterns of risk occurred by chance or represent true differences. Until more data are available, the frequency of endoscopic surveillance should be selected without regard to segment length.