Transatlantic Barrettʼs Esophagus: Are We Divided By More Than the ‘O’?

Abstract

Purpose: In January 2006, the British Society of Gastroenterology (BSG) updated their guidelines on the management of Barrett's Esophagus (BE). The most notable change cited in this publication was that intestinal metaplasia (IM) was no longer required within the columnar lined esophagus (CLE) to make the diagnosis of BE. Significantly, this made the UK definition far more inclusive than American College of Gastroenterology (ACG) definition of BE, throwing the future of transatlantic research in to doubt. Are we in the same ballpark? Methods: A retrospective analysis of all case notes and endoscopic records of patients diagnosed with BE after January 2006, as per the BSG guidelines, from a single UK tertiary referral centre was carried out. Those with a CLE but no IM (BSG diagnosis) were compared to those with CLE and IM (ACG diagnosis) and significant differences in the length of CLE, the number of biopsies taken and the presence of inflammation and hiatus hernia sought. Results: 196 patients were identified, all fulfilling the BSG diagnosis of BE; of those 76 had no detectable IM within the CLE, not diagnostic of BE by ACG criteria. Those patients meeting the ACG and not the BSG definition of BE had approximately 2.5 times more biopsies taken of their CLE (MWU, p=1.7 × 10-5) than those without IM within the CLE. Interestingly, the number of biopsies taken per cm was not significantly different between the two groups (MWU p=0.1838); patients with IM tended to have slightly fewer biopsies per cm (1.1/cm in the IM group vs. 1.2/cm in the non-IM group). The CLE was on average 1cm longer (MWU, p=0.043) in patients with IM within their CLE and they were more likely to have esophagitis (Chi-squared test p=0.016). No association was found between IM and the presence or length of an hiatus hernia, indicating correct anatomical sampling. Interestingly, those with IM were, on average, 7.5 years older than those without IM (p=.003). No gender difference detected between the two groups. Conclusion: Oscar Wilde was wrong, the US and UK are now separated by more than a common language; the former needs IM, the latter does not. From our data, is appears that strict adherence to the Seattle biopsy protocol in the UK would increase the diagnosis of IM. Consecutive CLE biopsies taken at a younger age in the US, however, may not detect IM, resulting in discharge from the BE screening program. In CLE less than 3 cm in length we suggest a minimal of 8 are biopsies taken; size does matter. However, in the absence of a unified diagnosis transatlantic research in BE is in jeopardy.