Stratification of the dysplasia and neoplasia risk using autofluorescence endoscopic surveillance of Barrett's esophagus.

BACKGROUND

This study assessed the efficacy of autofluorescence endoscopy (AFE) using the Onco-LIFE system and numerical color value (NCV) estimation in comparison to white light endoscopy (WLE) in endoscopic surveillance for identification of early dysplasia in Barrett's esophagus (BE) to aid in real-time image elucidation and minimize the overreliance on biopsy and histology.

METHODS

AFE, performed simultaneously during WLE, with biopsy was performed among 24 patients with BE. None of these patients had any obvious mucosal abnormalities in WLE. A total of 376 biopsies were taken, include 325 randomly collected according to Seattle Protocol and 51 additional biopsies, taken from the sites with pathological AF and NCV. All biopsy sites were assessed in vivo using WLE, AFE and NCV and compared to histological examinations, to estimate the efficacy of these methods in dysplasia assessment in BE.

RESULTS

In the case of 248 biopsies taken from sites with NCV below 1.0, two cases of unspecified dysplasia were recognized; in 14 biopsies with NCV above 2.0 in all cases the various grades of dysplasia were documented. Dysplasia was found in 42% of AFE + NCV- guided biopsy specimens, and in 7.1% of WLE-guided biopsy specimens. AFE + NCV detected high-grade dysplasia in 7 patients, 6 more than according to Seattle Protocol in WLE. The expected odds of dysplasia detection in a sample increases almost 1.9 times, if it was selected by the AFE method (p < 0.001), when compared to WLE and with accordance with Seattle Protocol guided biopsy.

CONCLUSION

The above results indicate that AFE + NCV using the Onco-LIFE system leads to improved BE lesion visualization for targeted biopsy with accurate histologic correlation compared to WLE and Seattle Protocol guided biopsy alone, and can serve to minimize additional biopsies.