Short segment columnar-lined oesophagus: an underestimated cancer risk? A large cohort study of the relationship between Barrett's columnar-lined oesophagus segment length and adenocarcinoma risk

Eur J Gastroenterol Hepatol. 2007 Nov;19(11):969-75. doi: 10.1097/MEG.0b013e3282c3aa14.

Abstract

Objective: Longer columnar-lined oesophagus (CLO) segments have been associated with higher cancer risk, but few studies have demonstrated a significant difference in neoplastic risk stratified by CLO segment length. This study establishes adenocarcinoma risk in CLO by segment length.

Methods: This is a multicentre retrospective observational study. Medical records of 1000 patients registered from six centres were examined and data extracted on demographic factors, endoscopic features and histopathology of oesophageal biopsies. Adenocarcinoma incidence was evaluated for patients stratified by their diagnostic segment length.

Results: Seven hundred and eighty-one patients had biopsy-proven CLO and a segment length recorded. Four hundred and ninety patients had at least 1 year of follow-up, providing 2620 patient-years of follow-up for incidence analysis. The overall annual adenocarcinoma incidence was 0.62%/year (95% confidence interval: 0.36-1.01). The annual incidence in the segment length groups was 0.59% (0.19-1.37) in short segment (<or=3 cm), 0.099% (0.025-0.55) in >3 <or=6 cm, 0.98% (0.27-2.52) in >6 <or=9 cm and 2.0% (0.73-4.35) in >9 cm; P=0.004.

Conclusion: This study demonstrates that the neoplastic risk of CLO varies according to segment length, and that overall, the risk of adenocarcinoma development is similar in short-segment and long-segment (>3 cm) CLO. The highest adenocarcinoma risk was found in the longest CLO segments and lowest risk in segments >3 <or=6 cm.

Publication types

  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adenocarcinoma / pathology*
  • Age Factors
  • Aged
  • Barrett Esophagus / pathology*
  • Biopsy
  • Chi-Square Distribution
  • Disease Progression
  • Esophageal Neoplasms / pathology*
  • Esophagoscopy
  • Esophagus / pathology*
  • Female
  • Follow-Up Studies
  • Humans
  • Incidence
  • Male
  • Middle Aged
  • Precancerous Conditions / pathology*
  • Regression Analysis
  • Retrospective Studies
  • Risk
  • United Kingdom