Introduction In the 2007 UK audit of acute upper gastrointestinal bleeding (AUGIB),1 44% of participant centres did not have facility for a dedicated bleeding rota. Gross measures of outcome, such as mortality, were found to be independent of the presence of a bleeding rota; however adherence to other aspects of guidance, such as time to endoscopy, in the absence of a bleeding rota, was not examined. We aimed to examine the performance in the absence of a dedicated bleeding rota, but with a “goodwill” out of hours service, against national guidelines and performance.
Methods Data were collected prospectively for all cases of AUGIB over a 2-month period from September 2010 to November 2010. Cases were identified from request forms in the endoscopy department. Data were collected for all endoscopies; repeat endoscopy on the same admission was recorded separately. Rockall score, if not included with the original endoscopy request was calculated retrospectively. Results were analysed in Excel. Statistics performed using Fischer's exact test.
Results There were 35 patients identified, referred for 40 procedures for AUGIB, 39 of which were carried out. Median age at presentation was 64 (35–92), 14 years younger than the national audit. The proportion of patients presenting with documented cirrhosis 23% (n=8) was significantly greater than in the national audit p=0.0026. Furthermore the proportion of patients with varices, although not necessarily responsible for the bleed, was greater 17% vs 8% p=NS. Rockall score was documented in 23% cases compared with 19% in the national audit. Rockall score was ≥3 in 73% compared with 51% in the national audit p=0.01. Time to endoscopy, regardless of Rockall score was <24 h in 72% and in patients with Rockall score ≥3 was 79% compared with 50% in both groups in the national audit. In the cases in our study for whom the Rockall score was documented, all had a Rockall score ≥3 and proceeded to endoscopy within 24 h. Mortality in both our study and the national audit was 10%.
Conclusion Our patient group was younger than the national average with a higher proportion with pre-endoscopy Rockall score ≥3. We demonstrate ongoing difficulties in encouraging documentation of risk assessment scores, however in our experience when the Rockall score was documented the endoscopy happened in a timely fashion. We demonstrate with data from our centre that early endoscopy is achievable in the majority of cases in the absence of a dedicated bleeding rota. This does not appear to affect overall mortality however risk assessment and early endoscopy are important as we strive to improve the service we offer our patients.
Competing interests None declared.
Reference 1. UK Comparative Audit of Upper Gastrointestinal Bleeding and the Use of Blood. London: British Society of Gastroenterology, 2007. http://www.bsg.org.uk/pdf_word_docs/blood_audit_report_07.pdf