The Neo-Squamocolumnar Junction after Radiofrequency Ablation (RFA): Do We Need to Biopsy if Normal Appearing on Endoscopy?

Abstract

Purpose: RFA is a safe and effective treatment for Barrett's esophagus (BE) with dysplasia and may decrease risk of progression to esophageal adenocarcinoma. Intestinal metaplasia (IM) of the gastro-esophageal junction (GEJ) has previously been shown to have a substantially lower cancer progression risk compared to BE. As a result, biopsy acquisition of an irregular z-line is discouraged. The proportion of patients with dysplasia or IM of the GEJ in the absence of endoscopic evidence of BE post-ablation is not known. AIM: We evaluated the proportion of post-RFA patients with IM or dysplasia of the GEJ who had no endoscopic evidence of BE. Methods: We reviewed all patients with BE and high grade dysplasia (HGD) who were referred for RFA at a tertiary care referral center between 3/2004 and 3/2012. The EGD reports of all patients following RFA were reviewed. Those patients who did not have any columnar metaplasia visualized in the esophagus and those with an irregular Z line, without tongues, islands or nodules, were assessed. Presence of an irregular z-line was defined as a columnar segment < 1 cm. The pathology from each GEJ biopsy was then reviewed for evidence of intestinal metaplasia or dysplasia. Results: Seventy three patients (63 male, age 64.3±10 years, BMI 29.9±5.4 kg/m2, all with entry histology of HGD and all without endoscopic evidence of BE) fulfilled the inclusion criteria. Patients were followed for a mean length of 25±19 months following initial RFA. Forty six patients (63%) had documented evidence of an irregular z-line on endoscopic evaluation. Fifteen patients (20.2%) were found to have either IM or dysplasia of the GEJ following a normal EGD post-RFA [5 (6.8%) with HGD, 5 (6.8%) with low-grade dysplasia, 1 (1.4%) indefinite for dysplasia and 4 (5.5%) with IM]. Univariate analysis did not show age (p=0.08), BMI (p=0.99), gender (p=0.96), number of RFA sessions (p=0.65) or presence of irregular z-line (p=0.36) to be associated with having an abnormal GEJ biopsy. On multivariate analysis, controlling for the presence of an irregular Z-line and number of RFA treatments; presence of hiatal hernia was found to increase the risk of IM or dysplasia of the GEJ [OR 1.45 (1.1-2.04), p=0.02]. Conclusion: In our cohort of patients, 20% had findings of dysplasia or IM of GEJ after RFA with no endoscopic evidence of BE. The rate of overall dysplasia (15%), especially HGD (7%), is higher than would be expected with IMGEJ patients (7% and 0%, respectively) found on population based screening by our group. Unlike IMGEJ in patients without a prior diagnosis of BE, we recommend routine biopsies of the GEJ in post-RFA patients with a normal appearing esophagus during surveillance endoscopy. Disclosure: Dr. Wang - Consultant: Covidien (Barrx), Ninepoint Medical Research, CDX Diagnostics, Pinnacle Biologics. Research: Covidien (Barrx), Fujinon, Ninepoint Medical. Lori Lutzke - Research: Covidien (Barrx), Fujinon, Ninepoint Medical. Dr. Iyer - Consultant: Takeda.