Acetic acid chromoendoscopy for detection of neoplastic Barrett’s oesophagus: stepped wedge cluster randomised controlled trial (ACID-trial)

Abstract

Abstract Objective To assess the added value of acetic acid chromoendoscopy (AAC) for the detection of dysplasia and oesophageal adenocarcinoma during Barrett’s oesophagus surveillance endoscopies. Design Prospective, open label, stepped wedge cluster, randomised trial. Setting 18 Dutch community hospitals. Participants Patients undergoing surveillance for Barrett’s oesophagus were eligible, excluding those with known or previously treated dysplasia. Interventions Hospitals randomly shifted from surveillance without AAC (control) to AAC (intervention), with endoscopists trained before implementation. Targeted biopsies were obtained if lesions were present; random four quadrant biopsies were conducted. Main outcome measures The primary endpoint was the neoplasia detection rate, defined as the percentage of participants with low grade dysplasia, high grade dysplasia, and oesophageal adenocarcinoma in any biopsy, confirmed by an experienced gastrointestinal pathologist. The secondary endpoint was the neoplasia detection rate in targeted biopsies. Results 2267 participants were included, of whom 1438 (63%) were allocated to surveillance without AAC and 829 (37%) to surveillance with AAC; AAC was applied in 679 (82%) allocated participants. Overall, neoplasia was diagnosed in 163 (7.2%) participants (111 (4.9%) low grade dysplasia, 30 (1.3%) high grade dysplasia, and 22 (1.0%) oesophageal adenocarcinoma). AAC use did not increase the neoplasia detection rate (7.4% (50/679) with AAC versus 7.1% (113/1588) without AAC; odds ratio 0.97, 95% confidence interval 0.65 to 1.44; adjusted for calendar time and Barrett’s oesophagus length). Suspected lesions were detected in 95 (14.0%) participants receiving AAC versus 101 (6.4%) without AAC (P<0.001), but the targeted neoplasia detection rate was not higher (2.1% (14/679) versus 1.8% (29/1588); odds ratio 1.05, 0.49 to 2.24). Conclusions Use of AAC did not improve detection of neoplasia in patients with Barrett’s oesophagus undergoing endoscopic surveillance in routine clinical practice. Trial registration www.onderzoekmetmensen.nl ; NL-OMON25059 (prospective registration; 4 Dec 2019).