448 Wide Area Transepithelial Sampling (WATS) Augments Detection of Barrett’s Esophagus and Associated Dysplasia in Patients With a Hiatal Hernia

Abstract

INTRODUCTION: Detection of BE and associated dysplasia is important prior to referral for endoscopic surveillance or treatment. Use of WATS with computer-assisted analysis, as an adjunct to forceps biopsy (FB), has been shown to markedly improve detection of both BE and dysplasia in both screening and surveillance cohorts (CDx Diagnostics, Suffern NY). Our aim was to assess the effect of a hiatal hernia (HH), a condition that leads to difficulty in performing routine forceps biopsies, on the utility of WATS in detection of BE and dysplasia. METHODS: Patients enrolled in IRB-approved trials where both FB and WATS were performed in the same endoscopy were eligible for inclusion. Cases were excluded if prior endoscopic ablation was documented. Data, including length of salmon-colored mucosa (SCM), HH size, FB and WATS results were aggregated and de-identified prior to analysis. RESULTS: A total of 4677 patients met study criteria (Table 1). For the entire cohort, the adjunctive yield for WATS in addition to FB was 111% for detection of any BE as well as associated dysplasia. Additional BE detection was less in the presence of a clinically significant hiatal hernia (HH) measuring at least 2 cm in axial length (116% vs 79% for any BE, P < 0.0001). However, augmented dysplasia detection remained constant at 111%. There was no significant change in detection of additional BE as hernia size increased in length (87% for HH< 2 cm, 81% for HH 2-4 cm, 72% for HH >4 cm). Of note, dysplasia detection rates increased with the length of SCM regardless of HH size (80% for < 1 cm, 120% for 1-3 cm, and 133% for 3+ cm), though there were insufficient dysplastic cases to reach statistical significance. Upstaging of non-dysplastic to dysplastic BE occurred when WATS was used regardless of HH size, with significantly higher rates for BE segments at least 3 cm in length (2% vs 14%, P = 0.0012). Similarly, the rate at which NDBE was upstaged to dysplastic BE by WATS was significantly higher when a HH was 2 cm or greater in length (2.17% vs 9.47%, P = 0.013). CONCLUSION: WATS improves the diagnostic yield of both BE and associated dysplasia above that achieved with FB, regardless of the presence or absence of a HH. Sampling the distal esophagus with FB above a HH can be technically challenging. WATS may provide the endoscopist with an easier and more effective means of confirming a BE diagnosis and identifying associated dysplasia.