Assessing Multiple Variables in Adenoma Detection: An Interim Analysis

Abstract

Introduction: To examine the impact of key variables such as anatomic abnormalities, patient age, and prep quality on colonoscopy withdrawal time and adenoma detection rate for screening exams. Methods: Approximately 2,000 paper and electronic records of colonoscopies from October 2012-October 2013 in an outpatient endoscopy center were manually reviewed. Screening colonoscopies on asymptomatic patients were included. Patients with recent endoscopy (>1 year), personal history of inflammatory bowel disease, familial adenomatous polyposis, or Lynch syndrome were excluded. Personal or family history of polyps was not exclusive. Variables including patient age, gender, number of polyps, polyp location, histology, withdrawl time (WT), prep and view quality, sedation dose, and performing endoscopist were recorded as continuous or categorical where appropriate. Data were analyzed using Microsoft Excel 2010 with Analysis ToolPak®. Chi-squared analysis, Student t test, linear regression, and ANOVA were applied to assess significance where appropriate. Results: One thousand-eight patients treated by 9 endoscopists met criteria for review. The mean age was 61.4±9.6 years; 53.3% patients were male and 47.7% were female. The mean WT was 13.4 minutes for all endoscopists (range: 9.7-18.4 minutes). The ADR was 37.8% among all endoscopists (range: 7-48%). Six hundred forty-eight (64.3%) patients were found to have polyps of any kind. Three hundred eighty-two (59.0%) of those patients had adenomatous polyps, and those with adenomas had 1.9 on average (SD 1.43). The presence of diverticulae, adhesions, or other noticeable anatomical abnormalities increased WT by 1.07 minutes (p=0.02), but did not affect ADR (p=0.34). Quality of bowel prep had no relationship with ADR (Chi-squared 0.18; DF 3; p=0.97), but was associated with a longer WT (12.8 minutes for excellent vs 16.8 minutes for fair or poor; p<0.01). Older age was associated with increased presence of adenomas (p<0.01). There was a trend toward longer WT with older age that did not reach statistical significance (p=0.23). Conclusion: Diverticulosis and other anatomic abnormalities tended to increase the time needed to adequately survey the colon. The quality of bowel prep may increase the time needed for adequate exam, but did not alter ADR in this series. Age was not independently associated with longer WTs. By increasing WT over the minimum accepted time, the ADR in this series is 37.8%, well over standard accepted rates of 25%. Disclosure - Dr. Randall - serves as a consultant to UCB, Bristol-Meyers Squibb, Takeda, Janssen, Salix, Abbvie, Amergen, Ironwood, Forest, Romark, Vecta, EnteraHealth, Boston Scientific, and Furiex. Dr. Taboada - nothing to disclose. Dr. Vizuete - nothing to disclose. Dr. Wendorf - nothing to disclose. Dr. Ayyar - nothing to disclose. Dr. Constantine - nothing to disclose.