DIMINUTIVE COLORECTAL POLYPS: THE LEARNING CURVE OF YOUNG ENDOSCOPISTS FOR THE CHARACTERIZATION OF LESIONS WITH I-SCAN TECHNOLOGY

Background and aim: Recently, Neumann et al.examine the learning curve of I-Scan technique for the prediction of hyperplastic and adenomatous colorectal lesion. However, there are limited data on interpretation of diminutive polyps (5 mm) by using i-Scan images. The diminutive polyps, as for as the majority of lesions detected during colonscopy, constitue the theme of the “predict, resect and discard” strategy. Given the size, they are the most difficult to interpret. The aim of this study was to evaluate the performance of young endoscopists for interpretation of hyperplastic and adenomatous diminutive colorectal polyps by using I-Scan images. Material and methods: Three young endoscopists with different practice experience and interpretation on i-scan images (0-1000 colonscopies performed) partecipated. At study entry, they underwent a teaching presentation of 25 pictures on pits and vascular patterns analysis of colorectal lesions visualized with i-scan technology. It followed a review of 90 images of diminutive polyps, divided into 2 sessions of 45 pictures each. All lesions were evaluated using iScan1 (SE5), iScan3 (SE4TEcCE2) and SE4TEp settings. At the end of the first module, the images were reviewed and histopathological diagnosis was known. The performance of young endoscopists was separately analyzed for each session and was calculated by using sensibility, specificity, LR+ and LR-. Results: Of the 90 lesions included, 47 polyps (52%) were <3 mm (median 3.6 mm, min/max 2/5 mm). The distribution for anatomic lesions was: rectum (n=25), left colon (n=24) and right colon (n=31). Forty-four polyps were hyperplastic and 46 adenomatous. After first session of 45 I-Scan images, the overall sensitivity, specificity, LR+ and LR- were 73%, 63%, 1.98 and 0.42, respectively. At the second module, the sensitivity, specificity, LR+ increases (80%, 73%, 2.96 respectively) and LR - decreases at 0.28. Conclusions: The young and less experienced endoscopists in I-Scan technologies can achieve an improvement for diminutive polyps characterization after reviewing cases previously analyzed. It is necessary to perform an adequate training period before reaching the results of skilled operators for the interpretation of polyps.