ThTP2.5 Diagnosis of intestinal metaplasia (IM) in the surveillance of Barrett’s oesophagus – a comparison between Cytosponge and endoscopic biopsy

Abstract

Abstract Introduction Regular endoscopy with oesophageal biopsies is recommended for the surveillance of Barrett’s oesophagus. Intestinal metaplasia (of all subtypes) carries the highest risk of progression to dysplasia and adenocarcinoma; thus, it is important for diagnosis of Barrett’s. Concerns exist about the accuracy of endoscopic diagnosis when obtaining biopsies and in ‘missing’ metaplasia between biopsies. Cytosponge is a novel, minimally invasive oesophageal cytology sampling device; this study assessed the correlation between endoscopic biopsies and Cytosponge cytology in the diagnosis of IM. Methods 192 patients with endoscopic Barrett’s oesophagus who were due for surveillance endoscopy in 2022 were identified and screened for suitability for Cytosponge. If suitable, patients were offered a Cytosponge. Cytosponge cytology results were compared with previous histopathological biopsy results. Results 127 patients (66%) had IM from original endoscopic biopsies. Many of these patients did not have images available for review to assess accuracy of original endoscopic diagnosis. Cytosponge detected IM in 88 cases (45%). Sensitivity was 67.77% (95% CI, 58.67%-75.98%) and specificity of 91.55% (95% CI, 82.51%-96.84%). The positive predictive value was 93.18% (95%CI, 86%-96%) and negative predictive value of 62.50% (95% CI, 56.04%-68.54%). Conclusion The accuracy of cytosponge is favourable and comparable to screening endoscopy. Cytosponge can be used as a screening tool and, potentially, as an alternative to endoscopy to identify IM. Further work is needed to determine accuracy of endoscopic diagnosis, given the high proportion (over one-third) of patients who are currently on a Barrett’s surveillance programme but do not have any histological diagnosis of IM.