Introduction Recent studies have provided conflicting data on whether colonoscopy lesion detection varies with time of day.1–3 We aimed to assess whether the time of procedure within a list affected polyp and adenoma detection rate (PDR, ADR), surrogate markers for colonoscopy quality, at a busy tertiary centre endoscopy unit in the UK.
Methods All patients undergoing colonoscopy in 2009 were retrospectively identified and included in the study. Patient demographics and colonoscopic findings were obtained from endoscopy and histology reports. Polyp detection rate, adenoma detection rate, polyps per colonoscopy (PPC) and adenomas per colonoscopy (APC) were calculated. Morning and afternoon lists were combined as both are 4 h long and results were analysed in 20 min segments. Statistical significance was determined by linear regression.
Results During the 12-month study period, 3923 colonoscopy procedures (60.4±15.3 years, 52.7% female) were performed by 42 endoscopists. 3718 procedures were available for analysis after excluding cases performed in the cross-over period between lists (1300–1400) and out of hours. The average PDR was 35.9% and the ADR 23.9%. Although PDR (p=0.27) and ADR (p=0.24) did not vary significantly during the course of a list, PPC (p=0.015, Abstract OC-048 figure 1) and APC (p=0.025) significantly reduced as lists went on. Results are shown in Abstract OC-048 table 1.
Conclusion The mean number of polyps and adenomas detected falls significantly during the course of a colonoscopy list. Notably, our data shows no difference in the traditional “detection rate” measures (ADR and PDR) and thus highlights that, in isolation, these may be inadequate measures of colonoscopy quality, and some measure of the number of adenomas detected is also required. The fall in number of adenomas per colonoscopy may relate to operator fatigue and implies that significant lesions could be missed at the end of a list. Further study is required to establish whether revising the length of endoscopy lists or providing a break during a list could prevent this drop off in performance.
Competing interests None declared.
References 1. Gurudu SR, et al. Am J Gastroenterol 2011;106:1466–71.
2. Lee A, et al. Gastrointest Endosc 2011;73:467–75.
3. Munson GW, et al. Gastrointest Endosc 2011;73:476–9.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.