Introduction The NHS bowel cancer screening programme (BCSP) commenced in 2006 and is one of the world’s largest organised screening programmes. Minimising adverse events (AEs) is an essential part of the programme. This study examines colonoscopic AEs and asociated risk factors.
Methods The study examined the AEs of bleeding, post-procedure pain and perforation on all colonoscopies conducted between Aug 2006 and Jan 2012 within the BCSP. Crude AE rates were reported and logistic regression used to examine the risk of an AE against various explanatory variables, including age, gender, polyp size, morphology, location and therapy device. Detailed analysis was conducted on the subset of single polypectomy procedures.
Results Data are from 130,831 colonoscopies including 167,208 polypectomies. There were 30,896 single polypectomy procedures. Caecal polyp location (but not the rest of the proximal colon) and increasing polyp size were the two most important factors predicting risk of bleeding, bleeding requiring transfusion, post-procedure pain and perforation. After adjustment for polyp size the RR for bleeding requiring transfusion for caecal diathermy snare polypectomy was 13.5 (95%CI 4.5–40.6) relative to the distal colon. After adjustment for polyp size the RR for perforation occurring during caecal non-pedunculated polypectomy was 7.7 (95%CI 1.3–46.1) relative to the distal colon. The predicted risk of bleeding requiring transfusion for a 50mm caecal polypectomy was as high as 1 in 22. Further analysis applying estimated risks from the single polyp analysis to multiple polypectomies showed that the absolute risk was not additive.
Conclusion This national study is the largest to focus on polyp-specific risk factors. Uniquely, we have been able to focus on colonoscopies where only one polypectomy was performed, allowing a detailed examination of risk to be undertaken where the characteristics of the polyp causing the AE can be clearly established. We have confirmed that the greatest risk factor for an AE is the size of the polyp.
Perhaps more importantly however, for the first time we have demonstrated a substantially and significantly increased risk for both bleeding requiring transfusion and perforation from caecal polyps (but not in other sites in the proximal colon) for a given polyp size, when compared to the distal colon. Endoscopists should take particular care in resecting caecal polyps.
Additional polypectomies, adjusted for polyp size and location, carry a much lesser risk and therefore confirm that performing multiple polypectomies during the same procedure maximises benefit whilst incurring a lesser level of harm per polyp resected.
Disclosure of Interest None Declared