Early Dysplasia After Initial Diagnosis of Barrett’s Esophagus: When Should We Initiate Surveillance?

Abstract

Introduction: Barrett’s Esophagus (BE) is a risk factor for developing esophageal adenocarcinoma (EAC). The rate of progression of BE to EAC increases in the presence of low grade dysplasia (LGD) and high grade dysplasia (HGD). However, the optimal time to initiate surveillance for dysplasia is not known. The American College of Gastroenterology guidelines recommend against starting dysplasia surveillance 1 year after initial diagnosis, a conditional recommendation that is supported by very low quality evidence. Our aim is to evaluate rates of early dysplasia after the initial diagnosis of Non-Dysplastic Barrett’s Esophagus (NDBE). Methods: We retrospectively analyzed a cohort of patients diagnosed with BE at a large tertiary care referral VA medical center. Endoscopic and histologic data were reviewed to determine the date of initial diagnosis of BE and the time to development of dysplasia. Patients who had LGD, HGD, or indeterminate for dysplasia (IFD) as the initial histological finding and those who had a history of BE prior to initial endoscopy were excluded. Early dysplasia was defined as LGD or HGD developing within 18 months of the initial diagnosis of NDBE. Results: Of 847 unique patients with BE, 97 with LGD or HGD were selected, of which 30 had an initial diagnosis of NDBE. The average age was 60.6 years, all were men, and 77% were Caucasian. 9 patients developed HGD. In this group 5 progressed early and 1 developed early IFD that would become HGD (6/9, 66.7%). 21 patients developed LGD. In this group 4 progressed early and 5 developed early IFD that would become LGD (9/21, 42.8%). The total number of patients who developed early dysplasia was 15 out of 30 (50%). The average time to early LGD, HGD, or IFD (that would become LGD or HGD) was 12.3 months. 6 patients (20%) had no endoscopic surveillance between the initial diagnosis of NDBE and that of LGD or HGD with an average time to dysplasia of 72.3 months. Conclusion: In a high risk cohort of NDBE patients, a significant proportion of those who developed dysplasia did so in the first 18 months. It is possible this represents an underestimate given that 20% of our cohort did not have appropriate endoscopic follow up between initial diagnosis and dysplasia. Thus, we find that a 1-year surveillance interval after initial diagnosis of NDBE in high-risk cohorts is warranted. Limitations of this study include its small sample size and retrospective design. Future large prospective studies are needed to confirm our findings.