Institutional Experience in Diagnostic Usability of Volumetric Laser Endo-Microscopy in Barrettʼs Esophagus

Abstract

Introduction: Barrett's Esophagus (BE) is a well-known precursor lesion to Esophageal Adenocarcinoma. Random four-quadrant biopsies every 1-2 cm (the Seattle Biopsy Protocol) in addition to aggressive sampling of any mucosal abnormality has been recommended; however, there is controversy about the intense biopsy sampling protocol in detecting the incidental Barrett's dysplasia and cancer. Volumetric laser endomicroscopy (VLE) is a novel second-generation optical coherence tomography imaging procedure that elicits volumetric cross-sectional and longitudinal images over six cm of the esophagus with resolution less than ten microns Methods: A prospective enrollment of patients with known BE at our institution was conducted from September 2014-Febuary 2015. Patients were undergoing surveillance and/or therapeutic endoscopic procedures along with VLE. Patients of age 18 and above with history of BE were included. Based on VLE findings; lesions with high-risk features were identified and targeted biopsies were taken. Seattle protocol biopsies, which is the standard of care, were also taken in four quadrants every 1 cm for surveillance. Biopsies were evaluated by two-blinded pathologists. Results: Fifty-two patients with known BE were included in the analysis. There were no adverse events associated with the use of VLE. Mean age was 66 +/- 10 years with 71% males (n=37). Out of 52 patients, 41(79 %) underwent prior radio frequency ablations and/or endoscopic mucosal resections. Seattle protocol biopsies were performed in all 52 cases and additional targeted biopsies were performed in 46 cases as prompted by findings seen on VLE. Out of 52 cases, 21 were found to have BE on histopathology. VLE scans were abnormal in 19 out of 21 patients giving the sensitivity of 91% and NPV of 75%. One patient with BE had normal HD-WLE but abnormal features were detected on VLE scan. The two patients missed on VLE scanning had small foci of non-dysplastic BE at the gastro-esophageal junction and these were appreciated as small islands on HD-WLE. All the cases of BE were detected with combination HD-WLE and VLE. Conclusion: VLE is a reliable clinical modality which can be used for detection of BE. When used in conjunction with HD-WLE, the diagnostic yield is favorable and can decrease the miss rate of BE. Our study demonstrated excellent sensitivity and NPV. We believe there is a learning curve and further experience will only improve the diagnostic yield.