Endoscopic tri-modal imaging for detection of early neoplasia in Barrett’s oesophagus: a multi-centre feasibility study using high-resolution endoscopy, autofluorescence imaging and narrow band imaging incorporated in one endoscopy system
- W L Curvers1,
- R Singh2,
- L-M Wong-Kee Song3,
- H C Wolfsen4,
- K Ragunath2,
- K Wang3,
- M B Wallace4,
- P Fockens1,
- J J G H M Bergman1
- 1Department of Gastroenterology and Hepatology, Academic Medical Centre, Amsterdam, The Netherlands
- 2Wolfson Digestive Disease Centre, Queen’s Medical Centre, Nottingham, UK
- 3Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA
- 4Division of Gastroenterology and Hepatology, Mayo Clinic, Jacksonville, Florida, USA
- Dr Jacques Bergman, Department of Gastroenterology and Hepatology, Academic Medical Centre, Meibergdreef 9, 1105 AZ, Amsterdam, The Netherlands;
- Revised 13 August 2007
- Accepted 28 August 2007
- Published Online First 26 October 2007
Objective: To investigate the diagnostic potential of endoscopic tri-modal imaging and the relative contribution of each imaging modality (i.e. high-resolution endoscopy (HRE), autofluorescence imaging (AFI) and narrow-band imaging (NBI)) for the detection of early neoplasia in Barrett’s oesophagus.
Design: Prospective multi-centre study.
Setting: Tertiary referral centres.
Patients: 84 Patients with Barrett’s oesophagus.
Interventions: The Barrett’s oesophagus was inspected with HRE followed by AFI. All lesions detected with HRE and/or AFI were subsequently inspected in detail by NBI for the presence of abnormal mucosal and/or microvascular patterns. Biopsies were obtained from all suspicious lesions for blinded histopathological assessment followed by random biopsies.
Main outcome measures: (1) Number of patients with early neoplasia diagnosed by HRE and AFI; (2) number of lesions with early neoplasia detected with HRE and AFI; and (3) reduction of false positive AFI findings after NBI.
Results: Per patient analysis: AFI identified all 16 patients with early neoplasia identified with HRE and detected an additional 11 patients with early neoplasia that were not identified with HRE. In three patients no abnormalities were seen but random biopsies revealed HGIN. After HRE inspection, AFI detected an additional 102 lesions; 19 contained HGIN/EC (false positive rate of AFI after HRE: 81%). Detailed inspection with NBI reduced this false positive rate to 26%.
Conclusions: In this international multi-centre study, the addition of AFI to HRE increased the detection of both the number of patients and the number of lesions with early neoplasia in patients with Barrett’s oesophagus. The false positive rate of AFI was reduced after detailed inspection with NBI.
Funding: The work of Wouter Curvers is supported by an unrestricted research grant from AstraZeneca BV, Zoetermeer, The Netherlands. The ETMI-prototype endoscopy equipment was provided by Olympus Inc., Tokyo, Japan.
Competing interests: None.