Probe-based confocal laser endomicroscopy: progress, challenges, and emerging applications

Probe-based confocal laser endomicroscopy (pCLE) provides real-time, cellular-level "optical biopsy" during endoscopy. This review synthesizes the technology, diagnostic performance, clinical use, safety, costs, and future directions of pCLE across gastrointestinal indications.

Targeted narrative review of randomized trials, meta-analyses, and observational studies on pCLE in Barrett's esophagus, gastric intestinal metaplasia (GIM), colorectal neoplasia, inflammatory bowel disease (IBD), and indeterminate biliary strictures. Outcomes included sensitivity/specificity, impact on biopsy yield and management, adverse events, and cost effectiveness.

pCLE combines a fiber-optic probe and IV fluorescein with 488-nm excitation to generate optical sections (~ 1 µm lateral resolution; depth ~ 55-70 µm). In Barrett's esophagus, adding pCLE to high-definition endoscopy nearly doubled neoplasia detection sensitivity versus standard protocols. For GIM, pooled sensitivity and specificity reached 97% and 94%. For colorectal lesions, sensitivity 81-91% and specificity 75-91% allowed in vivo characterization and fewer unnecessary resections. In IBD surveillance, pCLE identified more neoplasia with fewer biopsies (up to ~ 4.7-fold increase) and pooled sensitivity/specificity of 87-100%/90-94%. In indeterminate biliary strictures, pCLE integrated with ERCP/EUS improved accuracy and achieved negative predictive values up to 100%. Safety is favorable; risks relate mainly to fluorescein. Limitations include superficial penetration, narrow field of view, operator dependence, image-interpretation variability, procedure time, and high capital and per-case costs. Current guidelines view pCLE as an adjunct rather than routine standard. Emerging AI-assisted interpretation and molecular/ multispectral probes may standardize reads, expand indications, and improve yield.

pCLE strengthens endoscopic decision-making by enabling immediate, near-histologic assessment, targeted sampling, and earlier therapy. Broader adoption hinges on standardized training, validated image criteria, multicenter randomized trials with health-economic endpoints, and integration of AI and molecular imaging to reduce variability and cost.