Article Text

Download PDFPDF
PTU-080 Upper gastrointestinal bleeding: the role of a dedicated inpatient list
  1. M Hussein,
  2. C Podesta,
  3. M Carpani,
  4. K Tang,
  5. A Alisa,
  6. S Musa
  1. Gastroenterology, Royal Free London NHS trust, London, UK


Introduction United Kingdom guidelines recommend all upper gastrointestinal bleed (UGIB) patients have endoscopy within 24 hours (h). The 2015 National Confidential Enquiry into Patient Outcome and Death (NCEPOD) audit reported a rate of 65%. In addition to a well-established out of hours rota, a dedicated weekday afternoon inpatient endoscopy list was introduced at Barnet Hospital, Royal Free London NHS Trust, in 2016. We sought to assess the early impact of the service on UGIB patients.

Method A single centre retrospective study involving all inpatients with an UGIB at a large district general hospital (445 beds) serving a population >500,000, during a 5 month period (January-May 2016). The endoscopy procedure log was interrogated to identify UGIB patients, defined by haematemesis/coffee ground vomit, malaena and haemoglobin drop. Additional data were collected via electronic patient records and ‘Unisoft GI Reporting Tool’. Patient demographics, sedation dose, procedural time, endoscopy diagnosis, endoscopic intervention, time from presentation to endoscopy and hospital discharge within 24 hour of endoscopy, were evaluated.

Results In total, UGIB accounted for 152/439 (35%) of all inpatient endoscopies. Median age was 73 [interquartile range (IQR) 58–86], 53% were female. Median midazolam dose was 2 mg (IQR 1–3 mg) and fentanyl 50mcg (IQR 31.25-50mcg). Median procedure time was 15 min (IQR 10–20 min), 110/152 (73%) patients underwent endoscopy within 24 hour of presentation. Endoscopic findings included hiatus hernia 58/152 (38%), oesophagitis 41/152 (27%), gastritis 37/152 (24%), duodenitis 25/152 (16%), duodenal ulcer 17/152 (11%), portal hypertensive gastropathy 17/152 (11%), oesophageal varices 15/152 (10%), gastric ulcer 8/152 (5%), Mallory Weiss tear 2/152 (1%). Endoscopic therapeutic intervention was required in 21/152 (14%). Six (4%) patients had a normal endoscopy. 38/152 (25%) patients were discharged within 24 hour. Only 1/38 (2.6%) patient was re-admitted within 30 days. Overall 30 day mortality was 20/152 (13%).

Conclusion A dedicated inpatient endoscopy list can improve UGIB patient flow with a majority scoped within 24 hour and a quarter being discharged within 24 hour. Our rate was greater than reported by the NCEPOD audit. The timing of the list requires further validation as does the prospect of extending the service to achieve national recommendations.

Disclosure of Interest None Declared

  • Endoscopy
  • Upper GI bleed

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.