P296 Quality of reporting of lesions detected at surveillance colonoscopy for IBD
N. SHARMA1, A. Virk2, O. Nardone3, S. Smith3, P. Rimmer4, U. Shivaji3, S. Ghosh5, M. Iacucci5
1Institute of Immunology and Immunotherapy, University of Birmingham, Gastroenterology, Birmingham, UK, 2University of Birmingham, Gastroenterology, Birmingham, UK, 3Institute of Translational Medicine, Institute of Immunology and Immunotherapy, University of Birmingham, Birmingham, UK, 4Queen Elizabeth Hospital Birmingham, Gastroenterology, Birmingham, UK, 5Institute of Translational Medicine, Institute of Immunology and Immunotherapy- NIHR Biomedical Research Centre- University of Birmingham, Birmingham, UK
Background
International guidelines of ESGE and ASGE have laid out best practice for documentation of lesions at colonoscopy but few performance indicators have been proposed for surveillance colonoscopy in IBD. A recent publication has highlighted the key performance indicators for surveillance colonoscopy.1 We conducted an audit of current quality of colonoscopy reports documenting lesions detected during surveillance colonoscopy.
Methods
A retrospective analysis of patients who underwent colonoscopy for IBD surveillance over a five year period (2014–2019) at the Queen Elizabeth Hospital, Birmingham, UK was performed. The reports were analysed by independent academic doctors in the gastroenterology division trained in quality of endoscopic report analysis. Optimum criteria for documentation comprised lesion nature (Paris classification), size, documentation of Kudo classification and pit pattern, borders and ulceration.
Results
A total of 1028 colonoscopies were performed for IBD surveillance and the procedures were standardised with routine dye spraying since 2016. The mean patient age was 47.9 years (SD 16.8). Visual evidence of colonoscopic lesions was recorded in 273 cases. Key performance indicators documented for each endoscopic criterion and lesion nature is noted in the Table. Low-grade dysplasia was detected in 61 patients, and carcinoma in 4 patients; no patient had high-grade dysplasia. Benign lesions such as pseudopolyps were detected in the rest. 7 patients had sessile serrated lesions.
| Size | 198 (72.5%) |
| Paris classification | 132 (48.3%) |
| Kudo pit pattern | 124 (45.4) |
| Border | 27 (9.9) |
| Ulceration | 14 (5.1) |
Conclusion
At IBD surveillance colonoscopy, documentation of lesions is better for the domains of size, Paris classification and Kudo pit pattern, though not perfect. We also highlight that our colonoscopic documentation of borders and presence of ulcerations is done poorly. It is important that comprehensive training is undertaken to improve documentation as it is essential for the proper choice of management of these lesions.
Iacucci M