Diagnostic Accuracy of NICE Classification to Predict Deep Submucosal Invasion

The purpose of this study is to evaluate the diagnostic accuracy of NICE classification to predict deep submucosal invasion of colonic polyps \> 1 cm, considering histology as the gold standard, in a group of endoscopists who previously performed a training program.

In the first phase of the study, a learning program of NICE classification based on examples will be performed. Forty images will be evaluated before and after the learning program. In the second phase of the study, all consecutive patients who underwent to a colonoscopy will be included if a lesion greater than 1 cm is found and the endoscopy is performed with a high definition colonoscope with Narrow Band Imaging (NBI). Patients, lesions and endoscopy equipment characteristics will be recorded. Subsequently, histological diagnosis of the lesion will also be recorded. Evaluated test (NICE classification) was blinded to the pathologist who performed the gold standard test (histology). In case of adenocarcinomas, histological predictors of lymph node metastases, with predefined assessment criteria, will also be recorded. Finally, histological preparations of adenocarcinomas will be sent to the reference center and two experienced pathologists will examine them again.