PTH-016 Prevalence of post-discharge symptoms following colonoscopy and possible causal associations: quality assurance audit with telephone survey in 514 patients

Abstract

<h3>Introduction</h3> Focus on quality assurance in colonoscopy intensified as a prequel to the launch of the UK National Bowel Cancer Screening Programme. Previous research has examined auditable standards such as caecal intubation rate, adenoma detection and major complications but much less is known about the prevalence of post-discharge symptoms which may be self-limiting and not require readmission or formal medical input. <h3>Aim</h3> To determine the prevalence of post-discharge symptoms in patients that underwent colonoscopy and the possible causal associations. <h3>Methods</h3> As part of our Quality Assurance audit programme, a structured telephone survey was administered to 514 patients who had undergone colonoscopy (with air insufflation) at our University Hospital (July 07–Oct 08). The survey consisted of questions regarding overall experience, procedure comfort, intravenous cannula site problems, and post-discharge symptoms of; bleeding, nausea and vomiting, bloating and pain. Demographics, colonoscopy details and interview responses were collated. Characteristics of a “symptomatic group” (reporting any pain, bloating or n/v) (“SympGp”) were compared with “asymptomatic group” (“AsympGp”) and results expressed as either per cent or mean (SD). <h3>Results</h3> Overall prevalence: IV site symptoms: 6%; PR bleeding: 5.4%; Nausea or Vomiting: 0.6%; Bloating: 11%; Pain: 9.5%. 100 patients (19.5%) reported one or more symptoms of pain, bloating, nausea or vomiting (SympGp). Predictably, SympGp rated their overall experience (1–10 point scale: 10=best score) less favourably (7.5 (2.4) vs 8.1 (2.3), p=0.04) and their comfort during procedure (1–5 scale; 1=best score) was lower (2.4 (1.3) vs 1.9 (1.1), p=0.001). SympGp were more often female (55% vs 43.5%) and were younger (58 vs 64 years, p&lt;0.0001). However, the ASA score, trainee involvement, therapy rates and procedure times were not significantly different from the AsympGp (p&gt;0.05). Midazolam doses used were higher in the SymptGp (2.1 mg (1.3) vs 1.7 mg (1.2), p=0.03), but Fentanyl doses were similar (p=0.18). <h3>Conclusion</h3> Post-discharge symptoms are not uncommon with almost 1:16 experiencing cannula site symptoms, 1:20 bleeding and 1:5 symptoms of nausea/vomiting, pain or bloating. These latter patients had more procedure discomfort and less favourable overall experience despite larger doses of sedation. Younger, female patients were more likely to report post-discharge symptoms but trainee involvement, therapy, IV opiate administration and procedure time had no effect. Given the prevalence of these symptoms, repeating a similar audit using CO2 insufflation would be interesting.