Evaluation of polyp detection in relation to procedure time of screening or surveillance colonoscopy.

Abstract

Multiple factors influence the yield of colonoscopy for the detection of neoplasia. Few studies have addressed the impact of colonoscopy duration on procedure yield. The aim of our study was to determine whether endoscopist-specific procedure times correlate with the number and clinical significance of polyps detected at screening or surveillance colonoscopy. Procedural data from screening or surveillance colonoscopies performed at Mayo Clinic, Rochester MN, between January 1, 1996 and June 30, 2000, were reviewed. Individual endoscopists were characterized by their personal endoscopist procedure mean time (EPMT) to perform a negative colonoscopy. Procedure time included patient's consent and sedation. EPMT was then correlated with individual polyp detection rates. Overall, 10,159 colonoscopies were reviewed of which 4,312 (42.4%) yielded polyps. Polyp detection varied among endoscopists between 19.0% and 62.3%. There was a close correlation between EPMT and polyp yield (all sizes), r = 0.64, although correlation was weaker for polyps >10 mm (r = 0.42) and polyps >20 mm (r = 0.20). On multivariate analysis, longer mean endoscopist time was associated with colonic lesion detection, OR = 1.54 (95% CI 1.37-1.62). Longer mean procedure duration demonstrated a looser association with identification of polyps >10 mm, OR = 1.40 (1.19-1.64) and polyps >20 mm, OR = 1.03 (0.74-1.43). There is a direct correlation between colonoscopy procedure time and yield, with a three-fold variation of polyp detection rates. These results should prompt future prospective studies assessing the impact of colonoscopic withdrawal time on lesion detection.