Results from a Large Prospective Series of Acetic Acid Enhanced Chromoendoscopy in Evaluation of Barrett's Neoplasia: Is It Time to Change Our Practice?

Abstract

Introduction: The current practice of surveillance of Barrett's metaplasia entails collection of quadrantic biopsies at 2cm intervals. This is time consuming and generates significant work for the pathologist. Acetic acid sprayed on mucosa leads to reversible acetylation of nuclear proteins, improving the visualisation of the surface pattern and aiding the diagnosis of dysplasia/cancer. We aim to evaluate the role of chromoendocopy in the assessment of Barrett's neoplasia. Methods: We prospectively collected data on patients with Barrett's oesophagus who had undergone Chromoendoscopy between July 2004 and November 2008, performed by a single experienced endoscopist. The population consisted of surveillance cases and tertiary referrals requiring EMR. Gastroscopy was performed using fujinon (EG-590) gastroscopes and EPX 4400 processor. Patients were first examined with white light endoscopy. Acetic acid (2.0%) dye spray was used to evaluate the pit pattern of the Barretts mucosa. This was classified by the endoscopist as normal, mucosal neoplasia or invasive cancer (table 1). After chromoendoscopy, targeted biopsies of any lesions and quadrantic biopsies every 2 cm were taken. We compared the chromoendoscopic with the histological diagnosis to evaluate the usefulness of chromoendoscopy. Results: We had a total of 190 patients. The mean age was 67 (SD 11 range 36-87) with 75% male. The average length of Barrett's oesophagus was 4.7cm (SD 3.65 range 1-13) We found an excellent correlation between predicted histology on chromoendoscopy and the actual histology (r=0.98). We applied the Mcnemar's test to our data and found a significant difference between prechromoendoscopic diagnosis and post chromoendoscopy assisted histology (p=0.001).Chromoendoscopy had a false positive rate of 12% (23/190) and a false negative rate of 2% (4/190). 2/4 patients had multiple sessions of APC ablation for HGD in the past and were difficult to assess. The other 2/4 had LGD. Conclusions: This is the world's largest reported series of chromoendoscopic evaluation of Barrett's oesophagus. We found an excellent correlation between chromoscopic diagnosis and histology.Chromoendoscopy significantly improves the diagnosis of dysplasia and cancer in Barrett's oesophagus and the false negative rate is low. It has a potential to replace the policy of 2 cm quadrantic biopsies.