Colonoscopy Polyp Detection Rate is influenced by Timing of Colonoscopy.

Tze Yeong Teng, MD, Cheah Wei Keat, MD, FRCS, Cheryl Lau, MBBS, MRCS, MMed, FRCS, Edin. Jurong Health Services

Background: Colorectal cancer can be prevented by the detection and removal of adenomatous polyps. Studies suggest that polyp detection rate (PDR) may be affected by timing of colonoscopy due to endoscopist fatigue later in the day. The aim of our study is to assess the influence of the timing variables on PDR in a cohort of consecutive elective colonoscopies performed only by 4 certified endoscopists.

Methods: Univariate analysis and multivariate logistic regression analysis were performed on a prospective colonoscopy database, comparing PDR for colonoscopies performed in the morning shift (AM) and in the afternoon shift (PM) over a one-year period. Each shift lasted 4 hours. Only elective outpatient completed colonoscopies with adequate bowel preparation were included. Surveillance colonoscopies for cancers were excluded. PDR was defined as the detection of at least one histologically-confirmed polyp during colonoscopy.

Results: 585 colonoscopies were included, ADR was 26% in the cohort. Mean age was 59 (S.D 14). 306 (52.3%) were done in the AM and 279 (47.7%) were done in the PM. PDR was 30% in the AM group compared to 22% in the PM group (p=0.02). Excluding time needed for polypectomy, the mean time taken for scope withdrawal was significantly longer in the morning group (12min) compared to the afternoon group (9.9min) (p=0.002). Level of sedation, gender, previous abdominal surgery and indications for colonoscopy were not associated with PDRs. Factors found to be significant on univariate analysis were included in a multiple regression model to control for confounding. After controlling for co-variates such as individual endoscopists, bowel preparation quality and age, morning colonoscopies were associated with longer withdrawal time and led to significantly increased PDRs (OR 1.64, 95% C.I 1.02 – 2.61) p<0.04).

Conclusion: Timing of colonoscopy is an independent predictor for PDR. Colonoscopies performed in the morning have a longer mean withdrawal time, thus leading to a significantly higher PDR. As endoscopists' concentration decreases as the day progresses, this may account for the shorter time spent on colonoscopies on the afternoon.

« Return to SAGES 2015 abstract archive