456 Impact of Endoscopic Surveillance on Detection of Dysplasia Progression in Patients With Barrett's Esophagus Who Have Undergone Complete Eradication of Intestinal Metaplasia

Abstract

INTRODUCTION: Barrett's esophagus (BE) affects up to 15% of patients with gastroesophageal reflux disease and poses high risk of progression to esophageal adenocarcinoma (EAC). While radiofrequency ablation (RFA) and endoscopic mucosal resection (EMR) are effective treatments for dysplastic BE, risk of recurrence remains even after complete eradication of intestinal metaplasia (CEIM). Currently, endoscopic surveillance intervals after RFA ± EMR are informed only by expert opinion and weak evidence. It is unclear if current guidelines are cost-effective or suitable for all patients. The purpose of this study is to assess for variation in surveillance intervals and rates of recurrence in BE patients with high grade dysplasia (HGD). METHODS: A retrospective review of medical records was performed for 70 BE patients who underwent RFA ± EMR for biopsy-proven HGD between January 1, 2010 until November 30, 2018 and received at least 1 surveillance endoscopy with biopsies at 1 year after completion of treatment. These patients were analyzed for adherence to the ACG guidelines of an endoscopy every 3 months for the first year after treatment and for recurrence of dysplasia or development of EAC. RESULTS: At 1 year post treatment completion, CEIM was achieved in 48.6% of patients and non-dysplastic BE was achieved in 32.9% of patients. The average time between the first 4 surveillance intervals was 5.5 months. 14 of 70 patients were adherent to ACG guidelines, and 56 received endoscopies less frequently than recommended. Of the 14 patients who were adherent, 6 developed recurrent dysplasia (1 low grade dysplasia (LGD), 4 HGD) and 1 developed T1b carcinoma at one year. Of the 56 patients who received surveillance endoscopies less frequently than recommended, 7 developed recurrent dysplasia (4 LGD, 3 HGD) at one year. CONCLUSION:: Adhering to ACG surveillance guidelines following RFA±EMR treatment for HGD is difficult and requires close attention to the scheduling of these BE patients. Although our sample size is small, our study showed that the recurrence of dysplasia was not significantly reduced in individuals that were adherent to the ACG guidelines when compared to those that received surveillance less frequently than recommended. Further studies are needed to inform the development of evidence-based guidelines that identify recurrence or progression of dysplastic BE early, while minimizing the costs of over-surveillance.