In Vivo Histopathology for Detection of Gastrointestinal Neoplasia With a Portable, Confocal Miniprobe: An Examiner Blinded Analysis

Background & Aims: Confocal fluorescence microscopy (CFM) has been mentioned to be a promising tool for in vivo histology. Recently, a portable confocal miniprobe has been developed. Our aim was to evaluate the potential benefit of CFM for detection of gastrointestinal neoplasia. Methods: A total of 47 patients with known or suspected neoplasia in the upper (n = 34) or lower gastrointestinal tract (n = 13) were examined with standard endoscopes. After mucolyis with 5-10 mL of acetic acid 1.5%, chromoendoscopy with 2-5 mL cresyl violet 0.25% was performed, with the substance also being used as a fluorophore for CFM. Real-time video sequences were recorded. Thereafter, biopsies were taken or mucosectomy/polypectomy was performed from the same examined area. All stored sequences were put into a random order and assessed by a pathologist and a gastroenterologist both blinded to any data. Results: A total of 119 CFM video sequences were recorded of 85 benign or 34 neoplastic areas. Quality of CFM images was regarded too low in 24 (pathologist) and 14 sequences (gastroenterologist). For the pathologist, accuracy of CFM detecting neoplasia was 92.6% (suitable images) and 73.9% (intention to diagnose). The respective accuracy values for the gastroenterologist were 92.4% (suitable images) and 81.5% (intention to diagnose). Agreement between CFM and histopathology was excellent (kappa values, 0.821 and 0.817). Conclusions: We have demonstrated that CFM with a miniprobe has the potential to diagnose neoplasia during ongoing endoscopy. This system has the advantage that it can be used with standard endoscopes. Further studies are warranted for validation.