Variation In Polyp Detection Rates At Screening Colonoscopy

If a polyp is found, it can be removed immediately. The polyp is usually removed with small biopsy forceps or loop of wire (snare) that is advanced within a channel in the colonoscope.

Studies of screening colonoscopy have documented variation in adenoma (benign growth) detection rates between 10 percent and 50 percent. The clinical and medicolegal effects of missed lesions and the wide variation in polyp detection have created the need for quality standards for colonoscopy performance. The U.S. Multi-Society Task Force on Colorectal Cancer has established targets for continuous quality improvement. One such target is adenoma detection rates of 25 percent or greater for men and 15 percent or greater for women aged 50 years and older undergoing first-time screening colonoscopy. Another quality target is a mean minimum withdrawal time, currently suggested to be at least six minutes. With these targets in mind, researchers examined data from one of the longest ongoing programs of screening colonoscopy with the objectives of measuring the variation in polyp detection rates (PDRs) among endoscopists and identifying factors that account for the variation.

Researchers performed a retrospective cross-sectional analysis of summary-level data from endoscopists performing screening colonoscopy for the Lilly Colorectal Cancer Prevention Program in Indiana. All procedures were performed on average-risk, asymptomatic adults aged 50 years or older, who underwent first-time screening colonoscopy between September 1995 and June 2001. The study was approved by the institutional review board of Indiana University. The study included twenty-five endoscopists and 2,664 patients (1,108 women and 1,556 men). Summary-level data was analyzed for each endoscopist, including patient mean age, sex profile, mean procedure time, number of colonoscopies performed and the proportions of patients with any-sized polyp, any-sized adenoma, an adenoma one cm or larger in maximum diameter, and multiple (more than one) adenomas. Multiple linear regression analysis (a technique for determining the linear relationship between one dependent variable and two or more independent variables) identified factors that accounted for the variation in PDRs.

The mean procedure time (MPT) was 17.1 minutes. Adenoma detection rates ranged from seven percent to 44 percent and from zero percent to 13 percent for large polyps, which was not statistically significant. For all polyp categories, only one to three high outlier endoscopists (ie, higher than mean PDRs) were identified. Models that included the number of procedures, mean age, percentage of women, and MPT accounted for 36 percent to 56 percent of the variation in PDRs. Only MPT was significantly associated with PDRs across all models. Researchers concluded that PDRs vary widely among endoscopists, although only a few outliers (ie, endoscopists who had higher polyp detection rates) were identified, variation in PDRs was associated only with MPT and that further research is needed to determine the clinical importance of and reasons for this variation.

The researchers acknowledged several limitations including the use of summary-level data, which is more limited than individual data, indirect calculation of MPT, and whether each endoscopist's patient cohort was at comparable risk for colorectal neoplasia was uncertain.

In an accompanying editorial, Harminder Singh, MD, MPH, departments of internal medicine and community health sciences, University of Manitoba, Canada, and Gurkirpal Singh, MD, division of gasteroenterology and hepatology, Stanford University School of Medicine, Cal., stated that "whether the optimal withdrawal time to allow complete visualization of colonic mucosa for an average endoscopist should be six or 10 minutes or another time interval still needs to be better defined. Optimal withdrawal is essential to ensure complete colonic examination. Spending six or 10 minutes during withdrawal is a small price to pay for improving outcomes after colonoscopy."