Abstract
BACKGROUND AND AIMS: Recurrence of Barrett's esophagus (BE) and associated dysplasia is not uncommon after endoscopic eradication therapy (EET). However, the most optimal method for detection of residual/recurrent disease remains unknown. This study evaluated the utility of wide-area transepithelial sampling with 3-dimensional computer-assisted analysis (WATS-3D) as an adjunct to standard forceps biopsy (FB) sampling for detection of BE and dysplasia after EET. METHODS: Data were examined from 2 large commercial registries of community practices that prospectively evaluated BE surveillance using both standard FB sampling combined with WATS-3D. Comprising the study cohort were 1114 consecutive BE patients (mean age, 68.9; men 68.1%, women 31.9%) who had EET. The adjunctive (and absolute) increased yield of WATS-3D for detection of intestinal metaplasia (IM) and dysplasia was evaluated according to the extent of endoscopic residual BE. RESULTS: The WATS-3D adjunctive yield for detection of residual/recurrent IM or dysplasia was 52.5% and 91.5%, respectively. The highest adjunctive yield for detection of IM (260%) occurred in patients without any gross endoscopic evidence of residual or recurrent BE. The absolute yield for IM and dysplasia detection was 16% and 4.4%, respectively, and dysplasia detection was significantly greater in patients with visible BE compared with those with no endoscopic evidence of residual BE. Of 29 patients with high-grade dysplasia or esophageal adenocarcinoma detected by WATS-3D, FB sampling missed 11, including 7 where FB sampling did not detect any IM. The number of patients needed to test for detection of residual/recurrent BE was 6.2 and for detection of dysplasia was 22.9. CONCLUSIONS: WATS-3D is effective at increasing the diagnostic yield of IM and dysplasia in BE patients after EET when used as an adjunct to FB sampling.