Outcomes of patients with Barrett's oesophagus with low-grade dysplasia undergoing endoscopic surveillance in a tertiary centre: a retrospective cohort study

Background and Aim:Barrett's oesophagus predisposes individuals to oesophagealadenocarcinoma (OAC), with the risk of progression to malignancy increasing with thedegree of dysplasia, categorized as either low-grade dysplasia (LGD) or high-grade dys-plasia (HGD). The reported incidence of progression to OAC in LGD ranges from 0.02%to 11.43% per annum. In patients with LGD, Australian guidelines recommend 6-monthly endoscopic surveillance. We aimed to describe the surveillance practiceswithin a tertiary centre, and to determine the predictive value of surveillance as well asother risk factors for progression.Methods:Endoscopy and pathology databases were searched over a 10-year period tocollate all cases of Barrett's oesophagus with LGD. Medical records were reviewed todocument patient factors and endoscopic and histologic details. Because follow-uptimes varied greatly, survival analysis techniques were employed.Results:Fifty-nine patients were found to have LGD. Thirteen patients (22.0%) prog-ressed to either HGD or OAC (10 (16.9%) and three (5.1%) respectively); the annualincidence rates of progression to HGD/OAC and OAC were 5.5% and 1.1% respec-tively. All patients who developed OAC had non-guideline-adherent surveillance. ACox model found only two predictors of progression: (i) guideline-adherent surveil-lance, performed in 16 (27.1%), detected progression to HGD/OAC four times earlierthan non-guideline-adherent surveillance (95% confidence interval (CI)=1.3-12.3;P=0.016). (ii) The detection of visible lesions at exit endoscopy independentlypredicted progression (hazard ratio=6.5; 95% CI=1.9-22.8;P=0.003).Conclusion:Barrett's oesophagus with LGD poses a significant risk of progressionto HGD/OAC. Guideline-recommended surveillance is effective, but is difficult toadhere to. Clinical predictors for those who are more likely to progress are yet to bedefined