Adenoma Detection Rate is Not Influenced by the Timing of Colonoscopy

Abstract

Purpose: Multiple factors have been proven to affect adenoma detection rates especially adequacy of bowel preparation, withdrawal time, and the endoscopist's experience. Recently afternoon colonoscopies have been reported to be associated with lower adenoma detection rate (ADR) we conducted a study to determine if the timing of colonoscopy (morning vs. afternoon) affects adenomas detection rates (ADR) and polyps detection rates (PDR). Methods: We performed a retrospective review of outpatient screening colonoscopies completed between 2007 and 2009 by 14 gastroenterologists in our practice. Patient demographics, indications, timing, and findings of colonoscopy were noted. Mean withdrawal time, time to reach cecum, quality of preparation (graded from 1 to 5 according to Aronchik stool preparation scale), fellow involvement, experience of the endoscopist, detection of polyps, and pathology reports for adenoma detection were reviewed. ADR and PDR were calculated as detection of at least one adenoma and/or one polyp per colonoscopy respectively. Morning colonoscopies were defined as those which were performed before 12 noon and afternoon colonoscopies were performed after 12 noon. Results: A total of 1526 colonoscopies were included, of which 1048 (68.68%) colonoscopies were performed in morning and 478 (31.32%) in the afternoon. There was no difference in basic demographics between both morning and afternoon groups including age, gender, and ethnicity. A total of 575 patients had at least one polyp removed giving a PDR of 37.7%, of these, 329 were adenomas (ADR = 21.56%). The afternoon PDR (27.3%) was significantly lower than morning PDR (39.9%, P value = 0.008). More adenomas were detected in the morning than in the afternoon but the ADR was not statistically significant (22.8% vs. 18.8%; p=0.08). Mean withdrawal time in morning vs. afternoon was 9.12 vs. 9.25 min respectively (p=0.69). The mean ADR was higher in the group with withdrawal time >8 min but it was not statistically significant (p=0.34). There was no association between fellow involvement and polyp detection. The association between years of provider's experience and ADR was statistically significant (p=0.016). Conclusion: Timing of colonoscopy seems to be a determinant of polyp detection rate (PDR) but not adenoma detection rate (ADR). Though the ADR decreased as day progressed the change was not significant. The reasons and implications of this finding should be studied further.