PTH-053 High Definition (HD) Versus Standard Definition (SD) White Light Endoscopy for Detecting Early Neoplasia (EN) in Barrett’S Oesophagus (BO) During Surveillance Endoscopy. is It Time to Change the Standard of Care?

Introduction HD endoscopy systems provide superior image resolution. However, the utility of this new and more expensive technology in lesion detection has not been evaluated so far. Our aim was to assess whether using a HD endoscopy system translates to better outcomes compared to the SD system in terms of detecting EN in patients with BO. Methods The study included consecutive patients with non-dysplastic BO undergoing surveillance endoscopy between September 2008 and August 2012. Procedures were performed at the Nottingham University Hospitals NHS Trust using Olympus video endoscopy system (240 and 260 series with SD and HD gastroscopes and monitors) across two hospital sites. Data was retrieved from the hospitals’ endoscopy electronic database. Patients’ demographics, operator experience, endoscopy and histopathology findings were recorded. We excluded cases if other advanced imaging techniques were used or if cancer was diagnosed on index endoscopy. Logistic regression was performed to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CIs) comparing outcomes with HD and SD systems. Statistical models included the following potential confounders, chosen a priori based on the literature: number of biopsies taken; Male sex; trainee versus non-trainee endoscopist; HD versus SD system; BO length; and older age. Results The database search revealed 946 procedures, out of those, 425 were excluded. Data was analysed for the remaining 521 procedures (HD group n = 255, SD group n = 266). Age and sex distribution was similar for both groups. The HD system was superior to the SD system in the targeted detection of dysplastic lesions (aOR 3.27, 95%CI 1.27–8.40) as well as all dysplasia -random and target- (aOR 2.36, 95%CI 1.50–3.72). More false positive lesions (those with no dysplasia on target biopsies) were detected with the HD system (aOR 1.16, 95%CI 1.01–1.33) and it had a marginally higher yield of dysplasia on random biopsies alone (aOR 1.07, 95%CI 1.00–1.15). There was no benefit from the HD system in diagnosing all (random and target) high grade dysplasia (HGD) or cancers compared to SD endoscopy (aOR 0.93, 95%CI 0.83–1.04). Trainee endoscopists, number of biopsies and male sex were also associated with a statistically significant higher yield of dysplastic lesions. Conclusion The use of the HD endoscopy system is associated with better targeted and any dysplasia detection during surveillance endoscopies for BO and is independent of other factors. Endoscopists performing surveillance for BO should consider using HD endoscopes. Disclosure of Interest S. Sami: None Declared, V. Subramanian: None Declared, W. Butt: None Declared, G. Bejkar: None Declared, J. Coleman: None Declared, J. Mannath: None Declared, K. Ragunath Grant/Research Support from: Olympus (Keymed, UK).