Probe-based confocal laser endomicroscopy compared with standard four-quadrant biopsy for evaluation of neoplasia in Barrett's esophagus

Background and study aims: Surveillance of Barrett's esophagus includes endoscopic inspection with biopsy of suspicious lesions followed by four-quadrant biopsy of the remaining mucosa. We assessed the ability of probe-based confocal laser endomicroscopy (pCLE) to replace biopsy in the endoscopic evaluation of patients with Barrett's esophagus in a prospective and controlled setting. Patients and methods: A total of 68 patients who were referred for endoscopic assessment of Barrett's esophagus were included across three centers. pCLE recordings were interpreted live during the examination as well as in a blinded manner at least 3 months after endoscopy. pCLE diagnosis of neoplasia based on pre-defined criteria was compared with histopathology from suspicious as well as four-quadrant biopsies. Results: A total of 670 pairs of biopsies and pCLE video sequences were available for analysis, with neoplasia (high-grade dysplasia or cancer) being histologically diagnosed in 8.3%. Specificity and negative predictive value of pCLE in excluding neoplasia was 0.97 (90% CI 0.95-0.98) and 0.93 (0.91-0.95) for the blinded evaluation, and 0.95 (0.90-0.98) and 0.92 (0.90-0.94) for the on-site assessment. Positive predictive values (PPVs) and sensitivity were rather poor for both settings (46%/28% [blinded] and 18%/12% [on-site], respectively). Conclusions: pCLE can be regarded as non-inferior to endoscopic biopsy in excluding neoplasia of Barrett's esophagus mucosa. However, due to its low PPV and sensitivity, pCLE may currently not replace standard biopsy techniques for the diagnosis of Barrett's esophagus and associated neoplasia. Further technical development of pCLE and a better understanding of its role in relation to other imaging technologies are necessary.