Objectives: Endoscopic surveillance is recommended for patients with Barrett's esophagus (BE). However, it remains unclear if all BE patients benefit from long-term surveillance. We investigated the risk of esophageal adenocarcinoma (EAC) in BE patients in relation to number of successive endoscopies, years of follow-up, and calendar year.
Methods: We conducted a retrospective cohort study of male veterans with newly diagnosed BE during 2004-2009 with follow-up until 30 September 2011. EAC was verified using detailed structured electronic medical records reviews. We used Poisson regression to determine incidence rates, rate ratios (RR), and corresponding 95% confidence intervals (CI) for EAC according to number of successive endoscopies, years of follow-up independent of number of follow-up endoscopies, and calendar year of BE diagnosis.
Results: Among 28,561 male patients with BE, 406 developed EAC during 140,499 person-years of follow-up (median 4.9 years). EAC incidence rates increased with each additional endoscopy following a previous negative endoscopy (RR per additional endoscopy, 1.43; 95% CI, 1.25-1.64). Compared to the EAC incidence rate at the 1st follow-up EGD, the EAC incidence rate at the 5th follow-up EGD was ninefold higher (adjusted RR, 8.82; 95% CI, 4.90-15.9). EAC incidence was highest at the first year of follow-up (5.34 per 1,000 person-years); however, EAC rates starting from the second follow-up year increased during successive years of follow up. Compared to the EAC incidence rate in the 2nd year of follow-up, the EAC incidence rate was 1.5-fold higher in EGDs conducted ≥5 years after the index BE date (adjusted RR, 1.49; 95% CI, 1.07-2.10). In contrast, we found no significant change in EAC incidence rates by calendar year.
Conclusions: Persistence of non-neoplastic BE on multiple consecutive endoscopies was not associated with lower EAC risk. These findings argue against discontinuation of endoscopic surveillance in patients with persistent nondysplastic BE after multiple negative endoscopies.