Learning curve in mucosal healing (MH) and inflammatory activity assessment by using the Erlangen MH score for confocal laser endomicroscopy (eMHs) in inflammatory bowel diseases

Introduction: Confocal laser endomicroscopy (CLE) is a modern imaging technique that enables real time histology, during endoscopy. Monitoring histological activity is important in evaluating the therapy response in IBD. Our group has recently validated an endomicroscopic mucosal healing score (eMHs), which has to be interpreted by skilled endoscopists.

Aim & methods: Our first aim was to analyze the learning curve (LC) of MH assessment by CLE in endoscopists naïve to the CLE technique. Secondly, we comparatively investigated the LC between endoscopists and residents (i.e. acquainted neither to endoscopy nor to CLE).

Therefore, 4 study groups were established: a.) senior endoscopists (n = 3), board certified (> 2000 procedures, > 2 y of experience); b.) junior endoscopists (n = 3) (significant endoscopic skills, < 2 y of experience); c.) internal med. residents (n = 4) without endoscopic experience, and d.) a skilled endomicroscopist (n = 1). Initially, all attendees received a random set of 20 CLE images from 10 IBD patients with different inflammatory activity and a table with the eMHs (9 criteria) for a spontaneous assessment (based on histologic knowledge from medical school). Thereafter, all physicians participated in a training session (elementary lesions and assessment based on the eMHs). Subsequently, the same set of 20 CLE pictures was re-assessed (grouped per patient, before and after therapy). All physicians were blinded regarding patients' identity and disease activity. Assessment scores and duration from the pre- and post-teaching evaluation were statistically analyzed.

Results and conclusion: The average evaluation times before and after training for the groups a; b; c and d were: 25 vs. 12,33; 23 vs. 15; 24,5 vs. 15,25; 14 vs. 9 minutes, respectively). Overall, the evaluation time before instruction was significantly longer (p < 0.001) compared to the second. No significant differences were observed between the physicians with or without endoscopic experience (assessment duration and quality). Interobserver agreement in the groups (compared to the endomicroscopist) for group a; b and c were: 0,72; 0,52 and 0,75, respectively. In conclusion, the LC for MH and inflammatory activity assessment by CLE is fast and independent of endoscopic skills or experience.