PTU-110 Outcomes of inadequate bowel preparation colonoscopy: single-centre retrospective analysis. Issue 2 (June 2019)

• Record Type: Journal Article
• Authors: Siau, Keith; Carrothers, Kirsty; Shetty, Sharan
• Is Part Of: Gut. Volume 68:Issue 2(2019)
• Page Start: A49
• Page End: A49
• Publication Date: 2019-06
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2019-BSGAbstracts.99
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Quality in colonoscopy is underpinned by the quality of bowel preparation, however, data on real-world management following inadequate bowel preparation remain sparse. We aimed to audit the rates and outcomes of inadequate bowel preparation on colonoscopy, including subsequent investigations and rates of post-colonoscopy colorectal cancer (PCCRC). Methods: In this single-centre audit, all colonoscopies performed in 2015 were identified from an endoscopy reporting database. Patients with colonic resection were excluded. Bowel preparation was measured using the Aronchick scale and considered inadequate if reported as poor (<90% mucosal visualisation) despite washing. Retrospective follow-up was performed through electronic healthcare records with primary care linkage to identify subsequent investigations and rates of 3 yr-PCCRC. Subsequent investigation was defined as a repeat colonoscopy or relevant imaging within 6 months of an index procedure with inadequate preparation. Multivariable binary logistic regression was undertaken to identify predictors of subsequent investigation. Comparisons were performed at endoscopist-level and between adequate and inadequate bowel preparation groups using chi-square and Mann-Whitney tests. Results: Of 2305 procedures recorded over the study period, inadequate preparation was indicated in 235 (10.2%). This cohort was associated with lower caecal intubation rates (80.7% vs. 92.3%, P<0.001), higher polyp detection rates (47.7% vs 40.9%, P=0.04) but no differences in median age (65 vs. 64, P=0.190) or polyp resection rates (37.0% vs. 34.8%, P=0.484). Subsequent investigations were performed in 63 patients (26.8%), comprising repeat colonoscopy (N=31), CT imaging (N=31) and barium enema (N=1). Of the 18 endoscopists identified with >20 procedures in 2015, the rates of inadequate preparation ranged between endoscopists (0%-34.2%; P<0.001), although there was no significant difference in the practice of subsequent investigation (range 0–100%, P=0.511) [Abstract PTU110 figure 1]. On multivariable analysis, factors independently associated with subsequent investigations included: non-completion colonoscopy (OR 7.6, P<0.001), any abnormal diagnosis (OR 2.4, P=0.034), and younger age (P=0.040). No cases of 3 yr-PCCRC were identified within the cohort. Conclusion: Although the outcome of inadequate bowel preparation is an established quality metric in colonoscopy, its interpretation appears to vary between endoscopists, with the practice of instigating subsequent investigations also varying at procedural and patient level. Whilst our data should prompt reflection by local and national quality assurance groups, more robust studies are required to determine the impact of inadequate bowel preparation on patient outcomes such as PCCRC.