Introduction Despite advances in endoscopic haemostasis and pharmacological therapy, the mortality of Acute Upper Gastrointestinal Bleeding (AUGIB) in the UK remains high at 10%, increasing up to 26% for in-patient bleeds.1 Although achieving successful haemostasis is a key treatment goal, focusing management on prevention of secondary complications and comorbid illnesses may serve to reduce the mortality burden of the disease. We aimed to determine the specific cause of death in a large prospective UK audit of patients presenting with AUGIB.

Methods Analysis was performed on 6750 patients included in the 2007 UK Comparative Audit of Upper Gastrointestinal Bleeding and the Use of Blood.1 In this study prospective data were collected electronically on consecutive patients presenting to 208 UK hospitals with AUGIB between 1 May and 30 June 2007. In all patients whom died, clinicians were asked to comment whether they thought the death was due to haemorrhage or a related cause. In a separate question, they were then asked to list the mode of death, as listed as “Part 1A” on the patient's death certificate.

Results Of 6750 patients included in the final analysis, 675 (10%) died. Of the 675 patients who died, in 227 (33.6%) the clinician thought death was a direct result of gastrointestinal bleeding and in 448 (66.4%) it was from another cause. In terms of the mode of death as listed as “Part 1A” on the patient's death certificate, complete information was available for 474/675 (70.2%) patients. The majority of deaths (74%, 351/474) were non-bleeding related. The most common modes of death reported were respiratory failure, terminal malignancy, septicaemia and cardiac failure. The majority of patients who died from haemorrhage were acute admissions (66%, 81/123). There was no significant difference in the mean age of those who died as a direct result of bleeding, compared with those who died from another cause (73.8±15.9 vs 75.8±14.9). Similarly there was no significant difference in use of NSAIDS in those who died of a bleeding (7.5%) vs non-bleeding cause (6.5%).

Conclusion The majority of patients admitted with AUGIB did not die from further bleeding, but rather from another cause. Improving the outcome of AUGIB will require optimisation of treatment to reduce the risk of cardio-respiratory complications, sepsis and multi-organ failure rather than focusing only on achieving successful endoscopic haemostasis.