Abstract
BACKGROUND: The British Columbia Colon Screening Program (CSP) is a population-based program offering biennial fecal immunochemical test (FIT) to individuals age 50–74 years with follow-up colonoscopy for a positive FIT as well as individuals with a personal history of neoplastic polyps or a high risk family history of colorectal cancer. Over 20,000 colonoscopies are performed annually. In December 2018, program colonoscopists in Vancouver, Canada agreed to switch from a 2L polyethylene glycol (PEG) preparation to a 4L PEG preparation after studies suggested superiority of the higher volume preparation in achieving adequate bowel cleansing and improving adenoma detection rates (ADR). High quality bowel cleansing is critical to minimize repeat procedures and maximize neoplasia detection. AIMS: To compare the quality of bowel preparation and neoplasia detection rates using the 4L high volume split preparation (HVSP) versus the 2L low volume split preparation (LVSP) in patients undergoing colonoscopy in the BC CSP. METHODS: A retrospective review of consecutive patients undergoing colonoscopy through the CSP at St. Paul’s Hospital from Dec 2017-Apr 2018 and Dec 2018-Apr 2019 was conducted. Inclusion criteria included: age 50–74, patients undergoing colonoscopy for any reason through the BC CSP. Variables collected included: patient demographics, bowel preparation qualty and pathologic findings. ADR and sessile serrated polyp detection rate (SSDR) were analyzed. RESULTS: 462 colonoscopies were included, 280 in the LVSP group and 182 in the HVSP group. 8/280 (2.9%) had poor bowel preparation in the LVSP group, while 10/182 (5.5%) had poor bowel preparation in the HVSP group. The ADR and SSDR were 53.6% in LVSP vs. 50.0% in HVSP and 8.2% in LVSP vs. 8.8% in HVSP, respectively. CONCLUSIONS: In this preliminary evaluation, the high volume PEG-based split preparation did not reduce the proportion of inadequate bowel preparations. Further evaluation of a larger number of colonoscopies is planned. FUNDING AGENCIES: None