229 A Prospective, Randomized Trial of Adenoma Miss Rates At Colonoscopy Associated With 3-Minute vs. 6-Minute Withdrawal Time

Abstract

Introduction: A minimum withdrawal time for screening colonoscopy of 6 minutes has been proposed, based on an observational study. The 6-minute withdrawal time remains controversial and has not been compared in a standardized fashion to alternate withdrawal times. Aim: To perform a prospective, randomized trial to compare adenoma miss rates (AMR) associated with withdrawal times of 6 minutes vs. 3 minutes. Methods: Consecutive patients undergoing colonoscopy at Stanford University and Palo Alto Veterans Administration Hospital were randomized to undergo either a 3-minute or 6-minute withdrawal (1st pass), followed immediately by a tandem 6-minute withdrawal (2nd pass). AMR was defined as the number of adenomas detected on the 2nd pass divided by total number of adenomas found. A chi-square statistic was used to directly compare AMR. A z-statistic was used to determine effect of polyp size/location on AMR. A mixed-effects logistic regression model was then created to compare AMR, controlling for endoscopist, size/location of adenoma, and patient. Finally, a Cochrane-Armitage test for trend was used to evaluate the trend of adenoma detection associated with increasing withdrawal time. Results: 200 subjects were enrolled (99 in the 3-minute group, 101 in the 6-minute group). Demographics and indications for colonoscopy were similar between the two groups (Table 1). AMR was significantly higher in the 3-minute withdrawal group compared to the 6-minute withdrawal group (48.5% vs 21.8%; p≤0.0001). Adenomas were significantly more likely to be missed in the 3-minute withdrawal group in all 3 locations of the colon (right, transverse and left colon). AMR was higher in the 3-minute withdrawal group among adenomas that were ≤5mm; there was no significant difference in AMR for adenomas that were 6-9mm or ≥10mm. After controlling for endoscopist, size/location of the polyp/adenoma, and the patient, AMR remained significantly higher in the 3-minute withdrawal group compared to the 6-minute withdrawal group (OR 2.92, 95% CI=1.54-5.55) (Table 1). Adenoma detection rate (ADR) was similar between both 3 and 6 minute withdrawal groups (38.4% vs 40.6%, p=0.75); however, this study was not powered to detect differences in ADR. There was a significant trend towards higher total number of adenomas detected as withdrawal time increased (52 at 3 minutes, 87 at 6 minutes, 101 at 9 minutes, and 111 at 12 minutes; p<0.001) (Figure 1). Conclusions: A 3-minute withdrawal time leads to an unacceptably high AMR. A 6minute withdrawal time is more appropriate for colorectal cancer screening. Location did not affect AMR; however, adenomas that were ≤5mm were more likely to be missed in the 3-minute withdrawal group. Further research into the optimal withdrawal time, including evaluation of withdrawal times longer than 6 minutes, is warranted.