Experts are able to predict and differentiate between neoplastic and non-neoplastic colonic polyps with high accuracy and meet the PIVI (Preservation and Incorporation of Valuable Endoscopic Innovations) thresholds. However, this has not been replicated in non-experts and cannot be currently recommended in clinical practice. The aim of this randomised trial was to establish the optimum method of training to increase the diagnostic accuracy of diminutive/small colonic polyp histological prediction by trainee gastroenterologists.Methods
Gastroenterology trainees from the Midlands, UK, reviewed 78 videos (48 i-Scan-OE+30 NBI) of diminutive/small polyps. Participants were randomised to computer-based learning or didactic training. The 78 videos (randomised order) were re-assessed. The NICE (NBI International Colorectal Endoscopic) classification and SIMPLE (Simplified Identification Method for Polyp Labelling during Endoscopy) classification systems were used to classify diminutive/small polyps (figure 1).Results
16 trainees (12 gastroenterology trainees and 4 naïve endoscopists) were randomised to receive either self-training (n=8) or didactic training (n=8). A higher proportion of high confidence predictions of polyps were made by the self-training group vs didactic group when using the SIMPLE classification 77.1% vs. 69.9% (p<0.005) and using the NICE classification 77% vs. 69.8% (p=0.006). When using NICE classification, the sensitivity of the self-training group was 72% vs. 83% p<0.0005, and the accuracy 66.1% vs. 69.1%. When using SIMPLE classification the sensitivity was 78% vs. 83% (NS) and accuracy 65.7% vs. 69% (table 1).Conclusions
Self-learning for the prediction of diminutive/small polyp histology is a method of training that can achieve results similar to the more labour intensive and expensive didactic training. The availability of adequate self-learning teaching modules that teach how to differentiate neoplastic vs non-neoplastic colonic polyps with high accuracy could enable more widespread implementation of optical diagnosis in clinical practice.