High colonoscopic prevalence of proximal colon serrated polyps in average-risk men and women

Background: Proximal colon serrated polyps likely contribute to the decreased protection of colonoscopy against right-sided colorectal cancer.

Objective: To estimate the prevalence and extrapolated detection rate of proximal serrated polyps at screening colonoscopy.

Design: Retrospective study.

Patients: The study involved secondary analyses that used two databases. The first includes screening colonoscopies performed by 15 attending gastroenterologists at two academic endoscopy units between 2000 and 2009. The second includes average-risk patients who had previously participated in a randomized trial comparing high-definition chromocolonoscopy and white-light colonoscopy.

Main outcome measurements: Prevalence of proximal serrated polyps found by the highest-level detectors and proximal serrated polyp detection rates corresponding to adenoma detection rates of 25% in men and 15% in women, respectively.

Results: We analyzed 6681 procedures from the first database. Mean (± standard deviation) detection rates for adenomas and proximal serrated polyps were 38% ± 7.8% (range 17%-47%) and 13% ± 4.8% (1%-18%), respectively. There was a significant correlation between detection rates for adenomas and proximal serrated polyps for men (R = 0.71; P = .003) and women (R = 0.73; P = .002). Adenoma detection rates of 25% for men and 15% for women both corresponded to a detection rate of 4.5% for proximal serrated polyps. The prevalence of proximal serrated polyps found by the highest-level detector was 18%. The corresponding rate derived from the high-definition screening colonoscopy database was 20%.

Limitations: Retrospective study.

Conclusion: The prevalence of proximal colon serrated polyps in average-risk patients undergoing screening colonoscopy is higher than previously reported. An extrapolated proximal serrated polyp detection rate of 5% is suggested for average-risk men and women.