Systematic review with meta-analysis: Prevalence of prior and concurrent Barrett's oesophagus in oesophageal adenocarcinoma patients

Background The proportions of patients with oesophageal adenocarcinoma (OAC) diagnosed by Barrett's oesophagus surveillance or with pre-existing Barrett's oesophagus are unclear. Aim To estimate the prevalence of prior and concurrent Barrett's oesophagus diagnosis among patients with OAC or oesophagogastric junction adenocarcinomas (OGJAC). Methods We searched PubMed and Embase to identify studies published 1966-1/8/2020 that examined the prevalence of prior (>= 6 months) or concurrent Barrett's diagnosis (at cancer diagnosis) among OAC and OGJAC patients. Random effects models estimated overall and stratified pooled prevalence rates. Results A total of 69 studies, including 33 002 OAC patients (53 studies) and 2712 patients with OGJAC (28 studies) were included. The pooled prevalence of prior Barrett's oesophagus diagnosis in OAC was 11.8% (95% confidence interval [CI] 8.4%-15.6%). The prevalence of prior Barrett's oesophagus diagnosis was higher in single-centre resection studies (16.0%, 95% CI 8.7%-24.9%) than population-based cancer registry studies (8.4%, 95% CI 5.5%-11.9%). The prevalence of concurrent Barrett's oesophagus in OAC was 56.6% (95% CI 48.5%-64.6%). Studies with 100% early stage OAC had higher prevalence of concurrent Barrett's oesophagus (91.3%, 95% CI 82.4%-97.6%) than studies with <50% early OAC (39.7%, 95% CI 33.7%-45.9%). In OGJAC, the prevalence of prior and concurrent Barrett's oesophagus was 23.2% (95% CI 7.5%-44.0%) and 26.3% (95% CI 17.8%-35.7%), respectively. Conclusions Most patients with OAC have Barrett's oesophagus. Our meta-analysis found similar to 12% of OAC patients had prior Barrett's diagnosis, but concurrent Barrett's oesophagus was found in similar to 57% at the time of OAC diagnosis. This represents a considerable missed opportunity for Barrett's oesophagus screening.