324 Anesthesia Assistance in Screening Colonoscopy and Adenoma Detection Rate Among Trainees

Abstract

INTRODUCTION: The use of anesthesia assistance (AA) with propofol sedation for screening colonoscopy has been increasing over the past decade, raising concerns about possibly increased complication rates without a proven improvement in colonoscopy quality indicators like adenoma detection rate (ADR). The effect of AA on ADR has not been extensively studied in trainees learning colonoscopy. We aimed to determine if type of sedation used during screening colonoscopy affects trainee ADR. METHODS: Using the endoscopy databases of two hospitals in our medical center, we identified all screening colonoscopies performed by 15 trainees from July 2014 through June 2018, including all screening examinations where the cecum was reached. We recorded patient demographics, trainee ADR, year of training, mean withdrawal time, bowel preparation quality, supervising attending ADR (excluding procedures with trainees), and type of sedation used for each procedure. Multivariable analysis was used to determine factors independently associated with trainee ADR. RESULTS: We identified 1,420 unique patients who underwent screening colonoscopy by a trainee during the defined time period meeting the inclusion criteria. Of these, 459 (32.3%) were performed with AA. Overall trainee ADR was 39.6%, with ADR increasing from 35.0% in year one of training to 42.8% in year three (P = 0.047). ADR for cases with AA was 37.9% while ADR for conscious sedation (CS) cases was 32.0% (P = 0.374). Despite this absolute difference of 5.9%, the use of AA was not associated with finding an adenoma on colonoscopy on multivariable analysis when controlling for all covariables (OR 0.86; 95% CI 0.67-1.10). Factors independently associated with trainee ADR included male gender (OR 1.66; 95% CI 1.33, 2.08), patient age (OR for patients ages 70-79 compared to ages 50-59: 1.84 [1.33, 2.52]), optimal bowel preparation quality (OR 1.58; 95% CI 1.07, 2.33), and supervising attending ADR (OR for attending ADR 30-39% and ≥40% compared to attending ADR < 20%: 4.22 [1.58, 11.027] and 3.67 [1.37, 9.87], respectively). CONCLUSION: Despite providing the ability to more comfortably sedate patients for screening colonoscopy, potentially providing the increased time required to improve trainee ADR, we found that the use of AA for screening colonoscopy did not affect trainee ADR. These results on trainee ADR and sedation type suggest that the overall lack of association between AA use and colonoscopy quality indicators is applicable to the trainee setting.