106 Random Biopsies in Endoscopically Normal Esophagus Improve Yield of Recurrent Intestinal Metaplasia After Successful Radiofrequency Ablation

Abstract

Background: Radiofrequency ablation (RFA) treatment for dysplastic Barrett's Esophagus (BE) is associated with high rates of complete eradication of intestinal metaplasia (CEIM). However, recurrence of intestinal metaplasia (IM) in the esophagus is seen in approximately 25% of patients. The endoscopic findings associated with recurrence of IM are poorly described. Methods: We conducted a retrospective study of patients who underwent RFA for BE at University of North Carolina Hospitals between 2006 and 2013. Patients who achieved CEIM with at least two subsequent surveillance endoscopies were included. Patients were excluded if treated for non-dysplastic BE or invasive esophageal adenocarcinoma. Among patients with histologic evidence of recurrent BE during surveillance, we assessed the endoscopic findings associated with the recurrence. All patients had assessment of the distal esophagus by high-resolution white light and narrow band imaging, and all underwent regular biopsies according to a standard four-quadrant, q1cm procedure, as well as biopsy of endoscopically suspicious lesions. Endoscopic signs recorded included esophageal nodules (Paris classifications 0-Ip, 0-Is, 0-IIa), and areas suspicious for recurrent BE based on mucosal color changes. Statistical analysis was performed using SAS (version 9.3). Results: Of 302 patients, 178 met criteria for inclusion. These patients had 673 biopsy sessions (mean 3.8 sessions/pt). In total, 19 patients had histological recurrence of IM in the tubular esophagus (11%). Of these 19, only 5 (26%) had any endoscopic abnormality suggesting recurrence on endoscopy (table). The remaining 14/19 (74%, 95% confidence interval (CI): 49-91%) were found on routine surveillance biopsies. Of the 17 patients biopsied for a raised lesion or mucosal change suspicious for recurrent IM, only 5 (29%) actually had recurrence; 12 such biopsies were negative for recurrent IM. The median location for recurrent IM in targeted biopsies was 1 cm (Mean 2.5, S.D. 4.3) proximal to the top of the gastric folds; most (75%) specimens indicative of recurrence were from within 2 cm of the top of the gastric folds (figure). The odds ratio for recurrent disease in the setting of endoscopic signs was 17.7 (p < 0.001). Histologic grade was significantly higher for recurrence accompanied by endoscopic signs compared to those found on random biopsy (p = 0.016 for trend). Subsquamous recurrence was not identified in any biopsies regardless of endoscopic signs (95% CI: 0-23%). Conclusion: Histologic recurrence of IM following RFA was most common near the gastroesophageal junction. Subsquamous recurrence was not an important factor in recurrence. Most recurrences were found on routine, non-targeted biopsies, but endoscopic signs of recurrence including nodularity or apparent columnar-lined esophagus are associated with improved biopsy yield. Histology of Apparent and Non-Apparent Recurrence or Progression