Dedicated bleed units: should they be advocated?

BACKGROUND: Upper gastrointestinal haemorrhage (UGIH) is a common emergency, however, dedicated bleed units only exist in selected hospitals in the UK. OBJECTIVE: To evaluate the rebleeding and mortality rate of patients admitted with UGIH to a tertiary centre bleed unit in comparison with the current national standards and earlier unit performance in 1995-1998. METHODS: A retrospective case note review of demographics, the Rockall scores and final outcome was conducted for all patients admitted to the bleed unit over 24 months. RESULTS: Two hundred and fifty-five cases were identified with a mean age of 62 years and a median Rockall score of 3. Eighty-two percent of gastroscopies were performed within 24 h. Of these, 29% were undertaken after 5 p.m. Peptic ulcer and varices (15%) were the commonest diagnosis. The rebleeding (12%) and mortality rate (9%) were comparable with that of the units previous audit (P=0.47, 0.51, respectively) and the current national audit (P=0.58, 0.76, respectively). The number of patients requiring surgery has reduced from 6 to 0.4% in our unit over the last 8 years. Preendoscopy and postendoscopy Rockall scores were predictive of rebleeding (P=0.013, 0.045) and mortality (P=0.003, 0.01). CONCLUSION: This study has shown a consistently low rebleeding and mortality rate in patients with UGIH in a dedicated bleed unit. However there are limitations to the degree of improvements which can be demonstrated due to factors such as age and comorbidity.