Roel Bogie

Impact of endoscopist training on post-colonoscopy colorectal cancer rate

Introduction Ithasbeenclearlyshownthatpopulation-basedcolorectal cancer screeningeitherby fecal testing and/or colonoscopy significantly reduces CRC incidence and mortality. 1, 2 Although colonoscopy is considered as gold standard method for polyp detection and removal, it should be noted that colonoscopy is not perfect since still some CRCs are detected after a negative colonoscopy. 3 For a CRC screening program to be effective, the occurrence of these so-called post-colonoscopy colorectal cancers (PCCRCs) should be as low as possible. Several factors influence sensitivity of colonoscopy and thus occurrence of PCCRCs; for instance, image quality of the colonoscopes (high definition), bowel cleansing and skills of the endoscopist. 4 This way, the occurrence of PCCRCs is an important quality outcome parameter for CRC screening programs. It is known that PCCRCs occur due to various reasons (non-adherence to surveillance intervals, inadequate bowel examination, incomplete polyp resection, missed lesions or newly developed cancers), 5 where previous studies showed that a large subset of PCCRCs is due to missed lesions 6, 7 and could thereby be preventable. It could be hypothesized that precursor lesions of PCCRCs are often non-polypoid colorectal neoplasms (NP-CRNs) or serrated lesions, which are known to be mainly located in the proximal colon and have a flat appearance. 8, 9 This hypothesis is based on previous studies showing PCCRCs harbor these same characteristics; often located in the proximal colon, a flat appearance and smaller in size, 6, 10 making them more difficult to detect and resect. 3, 8 In addition, a subset of NP-CRNs and serrated lesions have different molecular features (compared to polypoid neoplasms), which may be more closely associated with carcinoma. 11, 12 Taken together, PCCRCs are likely to derive from non-polypoid/serrated lesions, which are more challenging to detect and resect. In order to prevent PCCRCs, training and advanced endoscopic skills are required. McGill et al. investigated the effect of endoscopy training in a longitudinal assessment (ranging from 200 to 1600 performed colonoscopies). 13 Over time, an increase in the detection of NP-CRNs was seen, indicating that non-polypoid adenoma detection is a skill that can be learned, but does require time and effort. 13 Up to now, no study has explored whether systematic training in detection and resection of NP-CRNs directly leads to reduction in PCCRC rate. Aim of the present study was to examine PCCRC rate and PCCRC etiology before and after implementation of a short systematic training program for the detection and resection of NP-CRNs.

9

Methods Data collection

At the Maastricht University Medical Center+, a specific training program for detection and resection of non-polypoid neoplasms was initiated in 2008. A specific aim of that training program was to increase awareness and detection of NP-CRNs. All endoscopists, faculty, and trainees at our university hospital were trained in a systematic training program comprising of lectures to improve awareness and basic knowledge, learning from experts, videos, cases and individual feedback. 14 From there, a prospective cohort study was initiated, including all colonoscopies performed between February 2008 and February 2012. Patients with hereditary CRC syndromes or history of

175