P243. Development and initial validation of a unique score for in vivo differentiation of ulcerative colitis and Crohn's disease featuring confocal laser endomicroscopy
G.E. Tontini1,2, J. Mudter3, M. Vieth4, R. Atreya1, C. Günther1, R. Kiesslich5, M. Vecchi2,6, M.F. Neurath1, H. Neumann1, 1University of Erlangen-Nuremberg, Medicine I, Erlangen, Germany, 2IRCCS Policlinico San Donato, Gastroenterology and Digestive Endoscopy Unit, San Donato Milanese, Italy, 3Sana Kliniken Ostholstein, Gastroenterology, Eutin, Germany, 4Klinikum Bayreuth, Institute of Pathology, Bayreuth, Germany, 5St. Marienkrankenhaus Katharina-Kasper, Medicine, Frankfurt am Main, Germany, 6University of Milan, Department of Biomedical Sciences for Health, Milan, Italy
Confocal Laser Endomicroscopy (CLE) allows on demand in vivo characterization of architectural and cellular tissue details during endoscopy. Recent evidences have shown that CLE can detect Crohn's disease (CD) and ulcerative colitis (UC) associated histological changes in vivo. Here, we prospectively assessed the efficacy of CLE for in vivo differentiation of IBD by developing a unique CLE scoring system based on histopathological hallmarks of colonic IBD involvement.
Consecutive patients with a well-established diagnosis of UC and CD underwent colonoscopy with fluoresceine-aided confocal imaging. In addition, biopsies were obtained for subsequent blinded histopathological analysis.
Overall, 79 patients (40 CD, 39 UC) were prospectively included. In CD, CLE showed significantly more often discontinuous inflammation (90% vs. 5%), focal cryptitis (75% vs. 13%) and discontinuous crypt architectural abnormality (90% vs. 5%). Conversely, UC was associated with severe, widespread crypt distortion (87% vs. 17%), decreased crypt density (79% vs. 22%) and frankly, irregular surface (90% vs. 17%). Significant differences were not seen for heavy, diffuse lamina propria cell increase or mucin preservation. Granulomas were not visible in any case. Based on these findings, we developed a scoring system for in vivo IBD differentiation based on endomicroscopic assessment (IDEA). Compared to the historical clinical diagnosis and histopathology, the IDEA score revealed excellent validity measures in both UC and CD subjects (sensitivity 97% and 90%, specificity 90% and 97%, positive predictive value 91% and 97%, negative predictive value 97% and 91%, accuracy 94% for both).
CLE enables in vivo characterization of most microscopic tissue features and inflammatory changes of tissue and cellular characteristics conventionally used by standard histopathology both to confirm diagnosis and to distinguish UC from CD.
However, according to the penetration depths of CLE, submucosal details or granulomas are not visible. The new scoring system (i.e. IDEA) allows for on demand in vivo differential diagnosis of UC and CD with high accuracy.