• 1. The number of patients and the number of lesions with HGD or early ACA detected with WLE and ETMI 2. The number of patients with HGD and early ACA detected with targeted biopsies only with ETMI and WLE [ Time Frame: 2 years ] [ Designated as safety issue: No ]
  

• 1. The positive predictive value (PPV) of HRE and AFI 2. The reduction of false-positive findings after NBI (both the initial in vivo NBI assessment as well as later assessment based on still images) [ Time Frame: 2 years ] [ Designated as safety issue: No ]
  

Patients with Barrett's esophagus are advised to undergo periodic endoscopic surveillance with random biopsies in an attempt to identify high-grade dysplasia (HGD) or early adenocarcinoma (ACA) at a time when intervention can be curative. This approach, however, can be time-consuming and is hindered by low sampling yield and random sampling error. Endoscopic Tri-Modal Imaging (ETMI) is a novel diagnostic modality that encompasses three advanced imaging features in one system: high-resolution endoscopy (HRE), autofluorescence imaging (AFI) and narrow-band imaging (NBI). HRE and AFI provide a bird's-eye view of 'red flag' areas which are then assessed by NBI for focused and more specific tissue characterization. The aim of this prospective, multi-center study is to compare the diagnostic performance of ETMI with that of standard white-light endoscopy (WLE) for identifying high-grade dysplasia (HGD) and early adenocarcinoma (ACA) in BE. A total of 84 BE patients will be recruited for the study and they will undergo both ETMI and WLE examinations in a randomized, crossover fashion. Standard surveillance biopsies and ETMI-targeted biopsies will be performed. The primary outcome will compare the number of patients and lesions with HGD or early ACA detected with WLE and ETMI. It is anticipated that ETMI will enhance the detection of high-grade Barrett's lesions relative to WLE.